diff --git "a/checkpoint-13391/trainer_state.json" "b/checkpoint-13391/trainer_state.json" new file mode 100644--- /dev/null +++ "b/checkpoint-13391/trainer_state.json" @@ -0,0 +1,5921 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.100219573400251, + "eval_steps": 479, + "global_step": 13391, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0075282308657465494, + "grad_norm": 87.12931823730469, + "learning_rate": 3.147218736930155e-07, + "loss": 12.3074, + "step": 48 + }, + { + "epoch": 0.015056461731493099, + "grad_norm": 64.15751647949219, + "learning_rate": 6.660393140945211e-07, + "loss": 15.7221, + "step": 96 + }, + { + "epoch": 0.02258469259723965, + "grad_norm": 81.39651489257812, + "learning_rate": 1.0173567544960265e-06, + "loss": 10.8027, + "step": 144 + }, + { + "epoch": 0.030112923462986198, + "grad_norm": 22.048904418945312, + "learning_rate": 1.3686741948975323e-06, + "loss": 8.9559, + "step": 192 + }, + { + "epoch": 0.037641154328732745, + "grad_norm": 11.542724609375, + "learning_rate": 1.7126725219573398e-06, + "loss": 8.8511, + "step": 240 + }, + { + "epoch": 0.0451693851944793, + "grad_norm": 10.625059127807617, + "learning_rate": 2.063989962358846e-06, + "loss": 9.3478, + "step": 288 + }, + { + "epoch": 0.05269761606022585, + "grad_norm": 14.21434211730957, + "learning_rate": 2.415307402760351e-06, + "loss": 8.8892, + "step": 336 + }, + { + "epoch": 0.060225846925972396, + "grad_norm": 13.216053009033203, + "learning_rate": 2.7666248431618565e-06, + "loss": 8.3008, + "step": 384 + }, + { + "epoch": 0.06775407779171895, + "grad_norm": 22.503334045410156, + "learning_rate": 3.117942283563362e-06, + "loss": 7.3455, + "step": 432 + }, + { + "epoch": 0.07512547051442911, + "eval_nli-pairs_loss": 6.591032028198242, + "eval_nli-pairs_runtime": 4.3469, + "eval_nli-pairs_samples_per_second": 46.01, + "eval_nli-pairs_steps_per_second": 1.15, + "eval_sts-test_pearson_cosine": 0.39488461174644296, + "eval_sts-test_pearson_dot": 0.15593446481859455, + "eval_sts-test_pearson_euclidean": 0.39975070029693277, + "eval_sts-test_pearson_manhattan": 0.4314268556737928, + "eval_sts-test_pearson_max": 0.4314268556737928, + "eval_sts-test_spearman_cosine": 0.3997824055251076, + "eval_sts-test_spearman_dot": 0.14324216739430146, + "eval_sts-test_spearman_euclidean": 0.40262274612650517, + "eval_sts-test_spearman_manhattan": 0.42925492969387746, + "eval_sts-test_spearman_max": 0.42925492969387746, + "step": 479 + }, + { + "epoch": 0.07512547051442911, + "eval_vitaminc-pairs_loss": 6.247874736785889, + "eval_vitaminc-pairs_runtime": 1.4382, + "eval_vitaminc-pairs_samples_per_second": 115.426, + "eval_vitaminc-pairs_steps_per_second": 2.781, + "step": 479 + }, + { + "epoch": 0.07512547051442911, + "eval_sts-label_loss": 3.371708869934082, + "eval_sts-label_runtime": 0.3925, + "eval_sts-label_samples_per_second": 509.603, + "eval_sts-label_steps_per_second": 12.74, + "step": 479 + }, + { + "epoch": 0.07512547051442911, + "eval_qnli-contrastive_loss": 3.5311310291290283, + "eval_qnli-contrastive_runtime": 0.2814, + "eval_qnli-contrastive_samples_per_second": 710.798, + "eval_qnli-contrastive_steps_per_second": 17.77, + "step": 479 + }, + { + "epoch": 0.07512547051442911, + "eval_scitail-pairs-qa_loss": 5.4017333984375, + "eval_scitail-pairs-qa_runtime": 1.1087, + "eval_scitail-pairs-qa_samples_per_second": 180.387, + "eval_scitail-pairs-qa_steps_per_second": 4.51, + "step": 479 + }, + { + "epoch": 0.07512547051442911, + "eval_scitail-pairs-pos_loss": 4.29502534866333, + "eval_scitail-pairs-pos_runtime": 2.4269, + "eval_scitail-pairs-pos_samples_per_second": 82.41, + "eval_scitail-pairs-pos_steps_per_second": 2.06, + "step": 479 + }, + { + "epoch": 0.07512547051442911, + "eval_xsum-pairs_loss": 3.540722370147705, + "eval_xsum-pairs_runtime": 1.0447, + "eval_xsum-pairs_samples_per_second": 191.444, + "eval_xsum-pairs_steps_per_second": 4.786, + "step": 479 + }, + { + "epoch": 0.07512547051442911, + "eval_compression-pairs_loss": 2.8060033321380615, + "eval_compression-pairs_runtime": 0.2399, + "eval_compression-pairs_samples_per_second": 833.621, + "eval_compression-pairs_steps_per_second": 20.841, + "step": 479 + }, + { + "epoch": 0.07512547051442911, + "eval_sciq_pairs_loss": 10.621454238891602, + "eval_sciq_pairs_runtime": 9.0638, + "eval_sciq_pairs_samples_per_second": 22.066, + "eval_sciq_pairs_steps_per_second": 0.552, + "step": 479 + }, + { + "epoch": 0.07512547051442911, + "eval_qasc_pairs_loss": 7.7197771072387695, + "eval_qasc_pairs_runtime": 1.2078, + "eval_qasc_pairs_samples_per_second": 165.595, + "eval_qasc_pairs_steps_per_second": 4.14, + "step": 479 + }, + { + "epoch": 0.07512547051442911, + "eval_openbookqa_pairs_loss": 7.620975494384766, + "eval_openbookqa_pairs_runtime": 1.053, + "eval_openbookqa_pairs_samples_per_second": 189.941, + "eval_openbookqa_pairs_steps_per_second": 4.749, + "step": 479 + }, + { + "epoch": 0.07512547051442911, + "eval_msmarco_pairs_loss": 8.353594779968262, + "eval_msmarco_pairs_runtime": 2.5338, + "eval_msmarco_pairs_samples_per_second": 78.932, + "eval_msmarco_pairs_steps_per_second": 1.973, + "step": 479 + }, + { + "epoch": 0.07512547051442911, + "eval_nq_pairs_loss": 7.995354652404785, + "eval_nq_pairs_runtime": 5.7107, + "eval_nq_pairs_samples_per_second": 35.022, + "eval_nq_pairs_steps_per_second": 0.876, + "step": 479 + }, + { + "epoch": 0.07512547051442911, + "eval_trivia_pairs_loss": 8.177907943725586, + "eval_trivia_pairs_runtime": 9.1824, + "eval_trivia_pairs_samples_per_second": 21.781, + "eval_trivia_pairs_steps_per_second": 0.545, + "step": 479 + }, + { + "epoch": 0.07512547051442911, + "eval_quora_pairs_loss": 1.1983369588851929, + "eval_quora_pairs_runtime": 0.659, + "eval_quora_pairs_samples_per_second": 303.48, + "eval_quora_pairs_steps_per_second": 7.587, + "step": 479 + }, + { + "epoch": 0.07512547051442911, + "eval_gooaq_pairs_loss": 7.4573974609375, + "eval_gooaq_pairs_runtime": 1.5917, + "eval_gooaq_pairs_samples_per_second": 125.65, + "eval_gooaq_pairs_steps_per_second": 3.141, + "step": 479 + }, + { + "epoch": 0.07512547051442911, + "eval_mrpc_pairs_loss": 2.1152825355529785, + "eval_mrpc_pairs_runtime": 0.241, + "eval_mrpc_pairs_samples_per_second": 829.751, + "eval_mrpc_pairs_steps_per_second": 20.744, + "step": 479 + }, + { + "epoch": 0.07528230865746549, + "grad_norm": 17.753456115722656, + "learning_rate": 3.469259723964868e-06, + "loss": 8.0369, + "step": 480 + }, + { + "epoch": 0.08281053952321205, + "grad_norm": 12.012594223022461, + "learning_rate": 3.820577164366374e-06, + "loss": 6.2732, + "step": 528 + }, + { + "epoch": 0.0903387703889586, + "grad_norm": 34.92698287963867, + "learning_rate": 4.1718946047678796e-06, + "loss": 7.8529, + "step": 576 + }, + { + "epoch": 0.09786700125470514, + "grad_norm": 23.564632415771484, + "learning_rate": 4.523212045169385e-06, + "loss": 5.8643, + "step": 624 + }, + { + "epoch": 0.1053952321204517, + "grad_norm": 22.126293182373047, + "learning_rate": 4.874529485570891e-06, + "loss": 6.3179, + "step": 672 + }, + { + "epoch": 0.11292346298619825, + "grad_norm": 25.067686080932617, + "learning_rate": 5.225846925972396e-06, + "loss": 6.1175, + "step": 720 + }, + { + "epoch": 0.12045169385194479, + "grad_norm": 29.170730590820312, + "learning_rate": 5.577164366373902e-06, + "loss": 5.2392, + "step": 768 + }, + { + "epoch": 0.12797992471769135, + "grad_norm": 29.377540588378906, + "learning_rate": 5.928481806775407e-06, + "loss": 5.8324, + "step": 816 + }, + { + "epoch": 0.1355081555834379, + "grad_norm": 33.512088775634766, + "learning_rate": 6.279799247176913e-06, + "loss": 5.1523, + "step": 864 + }, + { + "epoch": 0.14303638644918445, + "grad_norm": 32.54931640625, + "learning_rate": 6.6311166875784185e-06, + "loss": 6.0303, + "step": 912 + }, + { + "epoch": 0.15025094102885822, + "eval_nli-pairs_loss": 4.317643165588379, + "eval_nli-pairs_runtime": 4.0158, + "eval_nli-pairs_samples_per_second": 49.803, + "eval_nli-pairs_steps_per_second": 1.245, + "eval_sts-test_pearson_cosine": 0.6751726661173544, + "eval_sts-test_pearson_dot": 0.5308173325280101, + "eval_sts-test_pearson_euclidean": 0.6660519042507951, + "eval_sts-test_pearson_manhattan": 0.6723411683739887, + "eval_sts-test_pearson_max": 0.6751726661173544, + "eval_sts-test_spearman_cosine": 0.6458966208807124, + "eval_sts-test_spearman_dot": 0.5040208096497271, + "eval_sts-test_spearman_euclidean": 0.6447205374312966, + "eval_sts-test_spearman_manhattan": 0.6498560301461127, + "eval_sts-test_spearman_max": 0.6498560301461127, + "step": 958 + }, + { + "epoch": 0.15025094102885822, + "eval_vitaminc-pairs_loss": 6.306981563568115, + "eval_vitaminc-pairs_runtime": 1.4737, + "eval_vitaminc-pairs_samples_per_second": 112.643, + "eval_vitaminc-pairs_steps_per_second": 2.714, + "step": 958 + }, + { + "epoch": 0.15025094102885822, + "eval_sts-label_loss": 3.7677345275878906, + "eval_sts-label_runtime": 0.4136, + "eval_sts-label_samples_per_second": 483.59, + "eval_sts-label_steps_per_second": 12.09, + "step": 958 + }, + { + "epoch": 0.15025094102885822, + "eval_qnli-contrastive_loss": 2.8633975982666016, + "eval_qnli-contrastive_runtime": 0.2816, + "eval_qnli-contrastive_samples_per_second": 710.123, + "eval_qnli-contrastive_steps_per_second": 17.753, + "step": 958 + }, + { + "epoch": 0.15025094102885822, + "eval_scitail-pairs-qa_loss": 1.4301409721374512, + "eval_scitail-pairs-qa_runtime": 1.0526, + "eval_scitail-pairs-qa_samples_per_second": 190.003, + "eval_scitail-pairs-qa_steps_per_second": 4.75, + "step": 958 + }, + { + "epoch": 0.15025094102885822, + "eval_scitail-pairs-pos_loss": 2.468087673187256, + "eval_scitail-pairs-pos_runtime": 2.3275, + "eval_scitail-pairs-pos_samples_per_second": 85.928, + "eval_scitail-pairs-pos_steps_per_second": 2.148, + "step": 958 + }, + { + "epoch": 0.15025094102885822, + "eval_xsum-pairs_loss": 2.153658151626587, + "eval_xsum-pairs_runtime": 1.0409, + "eval_xsum-pairs_samples_per_second": 192.151, + "eval_xsum-pairs_steps_per_second": 4.804, + "step": 958 + }, + { + "epoch": 0.15025094102885822, + "eval_compression-pairs_loss": 1.6288033723831177, + "eval_compression-pairs_runtime": 0.2383, + "eval_compression-pairs_samples_per_second": 839.253, + "eval_compression-pairs_steps_per_second": 20.981, + "step": 958 + }, + { + "epoch": 0.15025094102885822, + "eval_sciq_pairs_loss": 9.78779411315918, + "eval_sciq_pairs_runtime": 9.0233, + "eval_sciq_pairs_samples_per_second": 22.165, + "eval_sciq_pairs_steps_per_second": 0.554, + "step": 958 + }, + { + "epoch": 0.15025094102885822, + "eval_qasc_pairs_loss": 3.7814972400665283, + "eval_qasc_pairs_runtime": 1.2108, + "eval_qasc_pairs_samples_per_second": 165.179, + "eval_qasc_pairs_steps_per_second": 4.129, + "step": 958 + }, + { + "epoch": 0.15025094102885822, + "eval_openbookqa_pairs_loss": 5.005772590637207, + "eval_openbookqa_pairs_runtime": 1.0415, + "eval_openbookqa_pairs_samples_per_second": 192.037, + "eval_openbookqa_pairs_steps_per_second": 4.801, + "step": 958 + }, + { + "epoch": 0.15025094102885822, + "eval_msmarco_pairs_loss": 4.574879169464111, + "eval_msmarco_pairs_runtime": 2.527, + "eval_msmarco_pairs_samples_per_second": 79.146, + "eval_msmarco_pairs_steps_per_second": 1.979, + "step": 958 + }, + { + "epoch": 0.15025094102885822, + "eval_nq_pairs_loss": 5.281248569488525, + "eval_nq_pairs_runtime": 5.6503, + "eval_nq_pairs_samples_per_second": 35.397, + "eval_nq_pairs_steps_per_second": 0.885, + "step": 958 + }, + { + "epoch": 0.15025094102885822, + "eval_trivia_pairs_loss": 4.913428783416748, + "eval_trivia_pairs_runtime": 9.0564, + "eval_trivia_pairs_samples_per_second": 22.084, + "eval_trivia_pairs_steps_per_second": 0.552, + "step": 958 + }, + { + "epoch": 0.15025094102885822, + "eval_quora_pairs_loss": 0.9212128520011902, + "eval_quora_pairs_runtime": 0.6293, + "eval_quora_pairs_samples_per_second": 317.839, + "eval_quora_pairs_steps_per_second": 7.946, + "step": 958 + }, + { + "epoch": 0.15025094102885822, + "eval_gooaq_pairs_loss": 3.932173490524292, + "eval_gooaq_pairs_runtime": 1.6066, + "eval_gooaq_pairs_samples_per_second": 124.483, + "eval_gooaq_pairs_steps_per_second": 3.112, + "step": 958 + }, + { + "epoch": 0.15025094102885822, + "eval_mrpc_pairs_loss": 1.0853501558303833, + "eval_mrpc_pairs_runtime": 0.2402, + "eval_mrpc_pairs_samples_per_second": 832.56, + "eval_mrpc_pairs_steps_per_second": 20.814, + "step": 958 + }, + { + "epoch": 0.15056461731493098, + "grad_norm": 28.37123680114746, + "learning_rate": 6.982434127979924e-06, + "loss": 5.7748, + "step": 960 + }, + { + "epoch": 0.15809284818067754, + "grad_norm": 26.708221435546875, + "learning_rate": 7.33375156838143e-06, + "loss": 4.8728, + "step": 1008 + }, + { + "epoch": 0.1656210790464241, + "grad_norm": 26.786447525024414, + "learning_rate": 7.685069008782934e-06, + "loss": 4.7375, + "step": 1056 + }, + { + "epoch": 0.17314930991217065, + "grad_norm": 26.215879440307617, + "learning_rate": 8.03638644918444e-06, + "loss": 4.6766, + "step": 1104 + }, + { + "epoch": 0.1806775407779172, + "grad_norm": 35.618831634521484, + "learning_rate": 8.387703889585947e-06, + "loss": 4.3209, + "step": 1152 + }, + { + "epoch": 0.18820577164366373, + "grad_norm": 37.166072845458984, + "learning_rate": 8.739021329987453e-06, + "loss": 3.7761, + "step": 1200 + }, + { + "epoch": 0.19573400250941028, + "grad_norm": 35.78367233276367, + "learning_rate": 9.090338770388957e-06, + "loss": 4.2161, + "step": 1248 + }, + { + "epoch": 0.20326223337515684, + "grad_norm": 36.299678802490234, + "learning_rate": 9.441656210790464e-06, + "loss": 4.9089, + "step": 1296 + }, + { + "epoch": 0.2107904642409034, + "grad_norm": 18.610933303833008, + "learning_rate": 9.792973651191968e-06, + "loss": 4.3406, + "step": 1344 + }, + { + "epoch": 0.21831869510664995, + "grad_norm": 9.592538833618164, + "learning_rate": 1.0144291091593475e-05, + "loss": 3.5664, + "step": 1392 + }, + { + "epoch": 0.22537641154328733, + "eval_nli-pairs_loss": 3.2245519161224365, + "eval_nli-pairs_runtime": 4.0436, + "eval_nli-pairs_samples_per_second": 49.461, + "eval_nli-pairs_steps_per_second": 1.237, + "eval_sts-test_pearson_cosine": 0.7002978854888552, + "eval_sts-test_pearson_dot": 0.5685392445320393, + "eval_sts-test_pearson_euclidean": 0.6963744527231541, + "eval_sts-test_pearson_manhattan": 0.7050517306003169, + "eval_sts-test_pearson_max": 0.7050517306003169, + "eval_sts-test_spearman_cosine": 0.6718756239728468, + "eval_sts-test_spearman_dot": 0.5416448961602434, + "eval_sts-test_spearman_euclidean": 0.6742379556154348, + "eval_sts-test_spearman_manhattan": 0.6824201536078427, + "eval_sts-test_spearman_max": 0.6824201536078427, + "step": 1437 + }, + { + "epoch": 0.22537641154328733, + "eval_vitaminc-pairs_loss": 6.715206623077393, + "eval_vitaminc-pairs_runtime": 1.4251, + "eval_vitaminc-pairs_samples_per_second": 116.481, + "eval_vitaminc-pairs_steps_per_second": 2.807, + "step": 1437 + }, + { + "epoch": 0.22537641154328733, + "eval_sts-label_loss": 4.016364097595215, + "eval_sts-label_runtime": 0.4049, + "eval_sts-label_samples_per_second": 493.95, + "eval_sts-label_steps_per_second": 12.349, + "step": 1437 + }, + { + "epoch": 0.22537641154328733, + "eval_qnli-contrastive_loss": 1.999517560005188, + "eval_qnli-contrastive_runtime": 0.2804, + "eval_qnli-contrastive_samples_per_second": 713.282, + "eval_qnli-contrastive_steps_per_second": 17.832, + "step": 1437 + }, + { + "epoch": 0.22537641154328733, + "eval_scitail-pairs-qa_loss": 1.0403239727020264, + "eval_scitail-pairs-qa_runtime": 1.0483, + "eval_scitail-pairs-qa_samples_per_second": 190.793, + "eval_scitail-pairs-qa_steps_per_second": 4.77, + "step": 1437 + }, + { + "epoch": 0.22537641154328733, + "eval_scitail-pairs-pos_loss": 1.9232473373413086, + "eval_scitail-pairs-pos_runtime": 2.3447, + "eval_scitail-pairs-pos_samples_per_second": 85.298, + "eval_scitail-pairs-pos_steps_per_second": 2.132, + "step": 1437 + }, + { + "epoch": 0.22537641154328733, + "eval_xsum-pairs_loss": 1.6821197271347046, + "eval_xsum-pairs_runtime": 1.0422, + "eval_xsum-pairs_samples_per_second": 191.901, + "eval_xsum-pairs_steps_per_second": 4.798, + "step": 1437 + }, + { + "epoch": 0.22537641154328733, + "eval_compression-pairs_loss": 1.1713249683380127, + "eval_compression-pairs_runtime": 0.2392, + "eval_compression-pairs_samples_per_second": 836.05, + "eval_compression-pairs_steps_per_second": 20.901, + "step": 1437 + }, + { + "epoch": 0.22537641154328733, + "eval_sciq_pairs_loss": 9.443825721740723, + "eval_sciq_pairs_runtime": 8.9916, + "eval_sciq_pairs_samples_per_second": 22.243, + "eval_sciq_pairs_steps_per_second": 0.556, + "step": 1437 + }, + { + "epoch": 0.22537641154328733, + "eval_qasc_pairs_loss": 2.9044029712677, + "eval_qasc_pairs_runtime": 1.2182, + "eval_qasc_pairs_samples_per_second": 164.182, + "eval_qasc_pairs_steps_per_second": 4.105, + "step": 1437 + }, + { + "epoch": 0.22537641154328733, + "eval_openbookqa_pairs_loss": 4.360418796539307, + "eval_openbookqa_pairs_runtime": 1.0522, + "eval_openbookqa_pairs_samples_per_second": 190.077, + "eval_openbookqa_pairs_steps_per_second": 4.752, + "step": 1437 + }, + { + "epoch": 0.22537641154328733, + "eval_msmarco_pairs_loss": 3.516049861907959, + "eval_msmarco_pairs_runtime": 2.5595, + "eval_msmarco_pairs_samples_per_second": 78.139, + "eval_msmarco_pairs_steps_per_second": 1.953, + "step": 1437 + }, + { + "epoch": 0.22537641154328733, + "eval_nq_pairs_loss": 4.016308784484863, + "eval_nq_pairs_runtime": 5.6561, + "eval_nq_pairs_samples_per_second": 35.36, + "eval_nq_pairs_steps_per_second": 0.884, + "step": 1437 + }, + { + "epoch": 0.22537641154328733, + "eval_trivia_pairs_loss": 3.781872272491455, + "eval_trivia_pairs_runtime": 9.0801, + "eval_trivia_pairs_samples_per_second": 22.026, + "eval_trivia_pairs_steps_per_second": 0.551, + "step": 1437 + }, + { + "epoch": 0.22537641154328733, + "eval_quora_pairs_loss": 0.8747495412826538, + "eval_quora_pairs_runtime": 0.6229, + "eval_quora_pairs_samples_per_second": 321.076, + "eval_quora_pairs_steps_per_second": 8.027, + "step": 1437 + }, + { + "epoch": 0.22537641154328733, + "eval_gooaq_pairs_loss": 3.0769765377044678, + "eval_gooaq_pairs_runtime": 1.552, + "eval_gooaq_pairs_samples_per_second": 128.863, + "eval_gooaq_pairs_steps_per_second": 3.222, + "step": 1437 + }, + { + "epoch": 0.22537641154328733, + "eval_mrpc_pairs_loss": 0.7370794415473938, + "eval_mrpc_pairs_runtime": 0.2401, + "eval_mrpc_pairs_samples_per_second": 832.867, + "eval_mrpc_pairs_steps_per_second": 20.822, + "step": 1437 + }, + { + "epoch": 0.2258469259723965, + "grad_norm": 29.73522186279297, + "learning_rate": 1.049560853199498e-05, + "loss": 4.7194, + "step": 1440 + }, + { + "epoch": 0.23337515683814303, + "grad_norm": 30.467117309570312, + "learning_rate": 1.0846925972396486e-05, + "loss": 3.6345, + "step": 1488 + }, + { + "epoch": 0.24090338770388958, + "grad_norm": 24.454021453857422, + "learning_rate": 1.1198243412797992e-05, + "loss": 3.5947, + "step": 1536 + }, + { + "epoch": 0.24843161856963614, + "grad_norm": 20.165475845336914, + "learning_rate": 1.1549560853199497e-05, + "loss": 4.0526, + "step": 1584 + }, + { + "epoch": 0.2559598494353827, + "grad_norm": 34.79319381713867, + "learning_rate": 1.1900878293601003e-05, + "loss": 3.7962, + "step": 1632 + }, + { + "epoch": 0.26348808030112925, + "grad_norm": 127.97925567626953, + "learning_rate": 1.2252195734002508e-05, + "loss": 4.1927, + "step": 1680 + }, + { + "epoch": 0.2710163111668758, + "grad_norm": 27.80243682861328, + "learning_rate": 1.2603513174404014e-05, + "loss": 3.6351, + "step": 1728 + }, + { + "epoch": 0.27854454203262236, + "grad_norm": 31.81105613708496, + "learning_rate": 1.295483061480552e-05, + "loss": 3.4256, + "step": 1776 + }, + { + "epoch": 0.2860727728983689, + "grad_norm": 32.932865142822266, + "learning_rate": 1.3306148055207025e-05, + "loss": 3.3175, + "step": 1824 + }, + { + "epoch": 0.2936010037641154, + "grad_norm": 31.197385787963867, + "learning_rate": 1.365746549560853e-05, + "loss": 3.4984, + "step": 1872 + }, + { + "epoch": 0.30050188205771644, + "eval_nli-pairs_loss": 2.852742910385132, + "eval_nli-pairs_runtime": 4.1529, + "eval_nli-pairs_samples_per_second": 48.16, + "eval_nli-pairs_steps_per_second": 1.204, + "eval_sts-test_pearson_cosine": 0.7132313507241694, + "eval_sts-test_pearson_dot": 0.559846529627866, + "eval_sts-test_pearson_euclidean": 0.7145939583366395, + "eval_sts-test_pearson_manhattan": 0.724552982808093, + "eval_sts-test_pearson_max": 0.724552982808093, + "eval_sts-test_spearman_cosine": 0.6912239915389706, + "eval_sts-test_spearman_dot": 0.5394217029355446, + "eval_sts-test_spearman_euclidean": 0.6946616748545426, + "eval_sts-test_spearman_manhattan": 0.70491424059339, + "eval_sts-test_spearman_max": 0.70491424059339, + "step": 1916 + }, + { + "epoch": 0.30050188205771644, + "eval_vitaminc-pairs_loss": 6.633151054382324, + "eval_vitaminc-pairs_runtime": 1.4454, + "eval_vitaminc-pairs_samples_per_second": 114.849, + "eval_vitaminc-pairs_steps_per_second": 2.767, + "step": 1916 + }, + { + "epoch": 0.30050188205771644, + "eval_sts-label_loss": 3.8717281818389893, + "eval_sts-label_runtime": 0.4016, + "eval_sts-label_samples_per_second": 498.049, + "eval_sts-label_steps_per_second": 12.451, + "step": 1916 + }, + { + "epoch": 0.30050188205771644, + "eval_qnli-contrastive_loss": 1.4170150756835938, + "eval_qnli-contrastive_runtime": 0.2814, + "eval_qnli-contrastive_samples_per_second": 710.85, + "eval_qnli-contrastive_steps_per_second": 17.771, + "step": 1916 + }, + { + "epoch": 0.30050188205771644, + "eval_scitail-pairs-qa_loss": 0.6900365948677063, + "eval_scitail-pairs-qa_runtime": 1.0611, + "eval_scitail-pairs-qa_samples_per_second": 188.492, + "eval_scitail-pairs-qa_steps_per_second": 4.712, + "step": 1916 + }, + { + "epoch": 0.30050188205771644, + "eval_scitail-pairs-pos_loss": 1.352358102798462, + "eval_scitail-pairs-pos_runtime": 2.3596, + "eval_scitail-pairs-pos_samples_per_second": 84.761, + "eval_scitail-pairs-pos_steps_per_second": 2.119, + "step": 1916 + }, + { + "epoch": 0.30050188205771644, + "eval_xsum-pairs_loss": 1.376610517501831, + "eval_xsum-pairs_runtime": 1.0393, + "eval_xsum-pairs_samples_per_second": 192.443, + "eval_xsum-pairs_steps_per_second": 4.811, + "step": 1916 + }, + { + "epoch": 0.30050188205771644, + "eval_compression-pairs_loss": 0.871735692024231, + "eval_compression-pairs_runtime": 0.2351, + "eval_compression-pairs_samples_per_second": 850.579, + "eval_compression-pairs_steps_per_second": 21.264, + "step": 1916 + }, + { + "epoch": 0.30050188205771644, + "eval_sciq_pairs_loss": 9.191713333129883, + "eval_sciq_pairs_runtime": 9.1572, + "eval_sciq_pairs_samples_per_second": 21.841, + "eval_sciq_pairs_steps_per_second": 0.546, + "step": 1916 + }, + { + "epoch": 0.30050188205771644, + "eval_qasc_pairs_loss": 2.369694709777832, + "eval_qasc_pairs_runtime": 1.2239, + "eval_qasc_pairs_samples_per_second": 163.415, + "eval_qasc_pairs_steps_per_second": 4.085, + "step": 1916 + }, + { + "epoch": 0.30050188205771644, + "eval_openbookqa_pairs_loss": 3.9601967334747314, + "eval_openbookqa_pairs_runtime": 1.0681, + "eval_openbookqa_pairs_samples_per_second": 187.247, + "eval_openbookqa_pairs_steps_per_second": 4.681, + "step": 1916 + }, + { + "epoch": 0.30050188205771644, + "eval_msmarco_pairs_loss": 3.0808801651000977, + "eval_msmarco_pairs_runtime": 2.5507, + "eval_msmarco_pairs_samples_per_second": 78.409, + "eval_msmarco_pairs_steps_per_second": 1.96, + "step": 1916 + }, + { + "epoch": 0.30050188205771644, + "eval_nq_pairs_loss": 3.4922549724578857, + "eval_nq_pairs_runtime": 5.7154, + "eval_nq_pairs_samples_per_second": 34.993, + "eval_nq_pairs_steps_per_second": 0.875, + "step": 1916 + }, + { + "epoch": 0.30050188205771644, + "eval_trivia_pairs_loss": 3.4910638332366943, + "eval_trivia_pairs_runtime": 9.1195, + "eval_trivia_pairs_samples_per_second": 21.931, + "eval_trivia_pairs_steps_per_second": 0.548, + "step": 1916 + }, + { + "epoch": 0.30050188205771644, + "eval_quora_pairs_loss": 0.833874523639679, + "eval_quora_pairs_runtime": 0.6419, + "eval_quora_pairs_samples_per_second": 311.554, + "eval_quora_pairs_steps_per_second": 7.789, + "step": 1916 + }, + { + "epoch": 0.30050188205771644, + "eval_gooaq_pairs_loss": 2.622526168823242, + "eval_gooaq_pairs_runtime": 1.5751, + "eval_gooaq_pairs_samples_per_second": 126.977, + "eval_gooaq_pairs_steps_per_second": 3.174, + "step": 1916 + }, + { + "epoch": 0.30050188205771644, + "eval_mrpc_pairs_loss": 0.4888114929199219, + "eval_mrpc_pairs_runtime": 0.2398, + "eval_mrpc_pairs_samples_per_second": 833.994, + "eval_mrpc_pairs_steps_per_second": 20.85, + "step": 1916 + }, + { + "epoch": 0.30112923462986196, + "grad_norm": 6.916851997375488, + "learning_rate": 1.4008782936010036e-05, + "loss": 2.7233, + "step": 1920 + }, + { + "epoch": 0.3086574654956085, + "grad_norm": 17.678085327148438, + "learning_rate": 1.4360100376411543e-05, + "loss": 3.6816, + "step": 1968 + }, + { + "epoch": 0.3161856963613551, + "grad_norm": 92.62138366699219, + "learning_rate": 1.4711417816813047e-05, + "loss": 3.3232, + "step": 2016 + }, + { + "epoch": 0.3237139272271016, + "grad_norm": 27.2542781829834, + "learning_rate": 1.5062735257214554e-05, + "loss": 3.3469, + "step": 2064 + }, + { + "epoch": 0.3312421580928482, + "grad_norm": 6.005978584289551, + "learning_rate": 1.5414052697616058e-05, + "loss": 3.7509, + "step": 2112 + }, + { + "epoch": 0.33877038895859474, + "grad_norm": 16.488624572753906, + "learning_rate": 1.5765370138017566e-05, + "loss": 3.1811, + "step": 2160 + }, + { + "epoch": 0.3462986198243413, + "grad_norm": 11.462204933166504, + "learning_rate": 1.611668757841907e-05, + "loss": 3.3341, + "step": 2208 + }, + { + "epoch": 0.35382685069008785, + "grad_norm": 33.485206604003906, + "learning_rate": 1.6468005018820577e-05, + "loss": 2.764, + "step": 2256 + }, + { + "epoch": 0.3613550815558344, + "grad_norm": 25.066240310668945, + "learning_rate": 1.681932245922208e-05, + "loss": 3.6488, + "step": 2304 + }, + { + "epoch": 0.36888331242158096, + "grad_norm": 28.305265426635742, + "learning_rate": 1.7170639899623588e-05, + "loss": 2.721, + "step": 2352 + }, + { + "epoch": 0.3756273525721455, + "eval_nli-pairs_loss": 2.527458667755127, + "eval_nli-pairs_runtime": 4.1153, + "eval_nli-pairs_samples_per_second": 48.599, + "eval_nli-pairs_steps_per_second": 1.215, + "eval_sts-test_pearson_cosine": 0.7258900302408404, + "eval_sts-test_pearson_dot": 0.5655223839113195, + "eval_sts-test_pearson_euclidean": 0.7228747263710285, + "eval_sts-test_pearson_manhattan": 0.732591374373909, + "eval_sts-test_pearson_max": 0.732591374373909, + "eval_sts-test_spearman_cosine": 0.707910346125958, + "eval_sts-test_spearman_dot": 0.5482635095738919, + "eval_sts-test_spearman_euclidean": 0.7064759533156177, + "eval_sts-test_spearman_manhattan": 0.7166423493246757, + "eval_sts-test_spearman_max": 0.7166423493246757, + "step": 2395 + }, + { + "epoch": 0.3756273525721455, + "eval_vitaminc-pairs_loss": 6.437549114227295, + "eval_vitaminc-pairs_runtime": 1.4278, + "eval_vitaminc-pairs_samples_per_second": 116.261, + "eval_vitaminc-pairs_steps_per_second": 2.801, + "step": 2395 + }, + { + "epoch": 0.3756273525721455, + "eval_sts-label_loss": 4.1980671882629395, + "eval_sts-label_runtime": 0.3956, + "eval_sts-label_samples_per_second": 505.555, + "eval_sts-label_steps_per_second": 12.639, + "step": 2395 + }, + { + "epoch": 0.3756273525721455, + "eval_qnli-contrastive_loss": 1.0682133436203003, + "eval_qnli-contrastive_runtime": 0.2789, + "eval_qnli-contrastive_samples_per_second": 717.152, + "eval_qnli-contrastive_steps_per_second": 17.929, + "step": 2395 + }, + { + "epoch": 0.3756273525721455, + "eval_scitail-pairs-qa_loss": 0.5046552419662476, + "eval_scitail-pairs-qa_runtime": 1.0451, + "eval_scitail-pairs-qa_samples_per_second": 191.365, + "eval_scitail-pairs-qa_steps_per_second": 4.784, + "step": 2395 + }, + { + "epoch": 0.3756273525721455, + "eval_scitail-pairs-pos_loss": 1.1998459100723267, + "eval_scitail-pairs-pos_runtime": 2.3442, + "eval_scitail-pairs-pos_samples_per_second": 85.316, + "eval_scitail-pairs-pos_steps_per_second": 2.133, + "step": 2395 + }, + { + "epoch": 0.3756273525721455, + "eval_xsum-pairs_loss": 1.1817097663879395, + "eval_xsum-pairs_runtime": 1.0372, + "eval_xsum-pairs_samples_per_second": 192.835, + "eval_xsum-pairs_steps_per_second": 4.821, + "step": 2395 + }, + { + "epoch": 0.3756273525721455, + "eval_compression-pairs_loss": 0.6974765062332153, + "eval_compression-pairs_runtime": 0.2369, + "eval_compression-pairs_samples_per_second": 844.401, + "eval_compression-pairs_steps_per_second": 21.11, + "step": 2395 + }, + { + "epoch": 0.3756273525721455, + "eval_sciq_pairs_loss": 8.970888137817383, + "eval_sciq_pairs_runtime": 9.0441, + "eval_sciq_pairs_samples_per_second": 22.114, + "eval_sciq_pairs_steps_per_second": 0.553, + "step": 2395 + }, + { + "epoch": 0.3756273525721455, + "eval_qasc_pairs_loss": 1.9235339164733887, + "eval_qasc_pairs_runtime": 1.2061, + "eval_qasc_pairs_samples_per_second": 165.828, + "eval_qasc_pairs_steps_per_second": 4.146, + "step": 2395 + }, + { + "epoch": 0.3756273525721455, + "eval_openbookqa_pairs_loss": 3.6225194931030273, + "eval_openbookqa_pairs_runtime": 1.0455, + "eval_openbookqa_pairs_samples_per_second": 191.296, + "eval_openbookqa_pairs_steps_per_second": 4.782, + "step": 2395 + }, + { + "epoch": 0.3756273525721455, + "eval_msmarco_pairs_loss": 2.664341926574707, + "eval_msmarco_pairs_runtime": 2.5305, + "eval_msmarco_pairs_samples_per_second": 79.036, + "eval_msmarco_pairs_steps_per_second": 1.976, + "step": 2395 + }, + { + "epoch": 0.3756273525721455, + "eval_nq_pairs_loss": 3.055206298828125, + "eval_nq_pairs_runtime": 5.6527, + "eval_nq_pairs_samples_per_second": 35.381, + "eval_nq_pairs_steps_per_second": 0.885, + "step": 2395 + }, + { + "epoch": 0.3756273525721455, + "eval_trivia_pairs_loss": 2.9497525691986084, + "eval_trivia_pairs_runtime": 9.0334, + "eval_trivia_pairs_samples_per_second": 22.14, + "eval_trivia_pairs_steps_per_second": 0.554, + "step": 2395 + }, + { + "epoch": 0.3756273525721455, + "eval_quora_pairs_loss": 0.7771684527397156, + "eval_quora_pairs_runtime": 0.626, + "eval_quora_pairs_samples_per_second": 319.495, + "eval_quora_pairs_steps_per_second": 7.987, + "step": 2395 + }, + { + "epoch": 0.3756273525721455, + "eval_gooaq_pairs_loss": 2.266879081726074, + "eval_gooaq_pairs_runtime": 1.5425, + "eval_gooaq_pairs_samples_per_second": 129.664, + "eval_gooaq_pairs_steps_per_second": 3.242, + "step": 2395 + }, + { + "epoch": 0.3756273525721455, + "eval_mrpc_pairs_loss": 0.36913084983825684, + "eval_mrpc_pairs_runtime": 0.2383, + "eval_mrpc_pairs_samples_per_second": 839.153, + "eval_mrpc_pairs_steps_per_second": 20.979, + "step": 2395 + }, + { + "epoch": 0.37641154328732745, + "grad_norm": 36.60768127441406, + "learning_rate": 1.752195734002509e-05, + "loss": 3.3609, + "step": 2400 + }, + { + "epoch": 0.383939774153074, + "grad_norm": 24.15782928466797, + "learning_rate": 1.7873274780426595e-05, + "loss": 2.6252, + "step": 2448 + }, + { + "epoch": 0.39146800501882056, + "grad_norm": 7.509932041168213, + "learning_rate": 1.8224592220828106e-05, + "loss": 3.5142, + "step": 2496 + }, + { + "epoch": 0.3989962358845671, + "grad_norm": 29.380950927734375, + "learning_rate": 1.857590966122961e-05, + "loss": 4.0597, + "step": 2544 + }, + { + "epoch": 0.4065244667503137, + "grad_norm": 28.593975067138672, + "learning_rate": 1.8927227101631114e-05, + "loss": 2.8512, + "step": 2592 + }, + { + "epoch": 0.41405269761606023, + "grad_norm": 21.228628158569336, + "learning_rate": 1.927854454203262e-05, + "loss": 2.717, + "step": 2640 + }, + { + "epoch": 0.4215809284818068, + "grad_norm": 43.00386047363281, + "learning_rate": 1.962986198243413e-05, + "loss": 3.4717, + "step": 2688 + }, + { + "epoch": 0.42910915934755334, + "grad_norm": 25.004785537719727, + "learning_rate": 1.9981179422835632e-05, + "loss": 3.1105, + "step": 2736 + }, + { + "epoch": 0.4366373902132999, + "grad_norm": 7.555154323577881, + "learning_rate": 2.0332496863237136e-05, + "loss": 2.7798, + "step": 2784 + }, + { + "epoch": 0.44416562107904645, + "grad_norm": 30.839733123779297, + "learning_rate": 2.0683814303638643e-05, + "loss": 3.3606, + "step": 2832 + }, + { + "epoch": 0.45075282308657466, + "eval_nli-pairs_loss": 2.284590721130371, + "eval_nli-pairs_runtime": 4.0714, + "eval_nli-pairs_samples_per_second": 49.123, + "eval_nli-pairs_steps_per_second": 1.228, + "eval_sts-test_pearson_cosine": 0.7382507781851606, + "eval_sts-test_pearson_dot": 0.5710221319397019, + "eval_sts-test_pearson_euclidean": 0.7307583601561211, + "eval_sts-test_pearson_manhattan": 0.7394202696141936, + "eval_sts-test_pearson_max": 0.7394202696141936, + "eval_sts-test_spearman_cosine": 0.7211579109789371, + "eval_sts-test_spearman_dot": 0.5515579746967598, + "eval_sts-test_spearman_euclidean": 0.7142073811971875, + "eval_sts-test_spearman_manhattan": 0.7240537218564107, + "eval_sts-test_spearman_max": 0.7240537218564107, + "step": 2874 + }, + { + "epoch": 0.45075282308657466, + "eval_vitaminc-pairs_loss": 6.35264253616333, + "eval_vitaminc-pairs_runtime": 1.4349, + "eval_vitaminc-pairs_samples_per_second": 115.687, + "eval_vitaminc-pairs_steps_per_second": 2.788, + "step": 2874 + }, + { + "epoch": 0.45075282308657466, + "eval_sts-label_loss": 4.186042308807373, + "eval_sts-label_runtime": 0.3983, + "eval_sts-label_samples_per_second": 502.129, + "eval_sts-label_steps_per_second": 12.553, + "step": 2874 + }, + { + "epoch": 0.45075282308657466, + "eval_qnli-contrastive_loss": 0.781445324420929, + "eval_qnli-contrastive_runtime": 0.2765, + "eval_qnli-contrastive_samples_per_second": 723.448, + "eval_qnli-contrastive_steps_per_second": 18.086, + "step": 2874 + }, + { + "epoch": 0.45075282308657466, + "eval_scitail-pairs-qa_loss": 0.4217279851436615, + "eval_scitail-pairs-qa_runtime": 1.0438, + "eval_scitail-pairs-qa_samples_per_second": 191.612, + "eval_scitail-pairs-qa_steps_per_second": 4.79, + "step": 2874 + }, + { + "epoch": 0.45075282308657466, + "eval_scitail-pairs-pos_loss": 1.051362156867981, + "eval_scitail-pairs-pos_runtime": 2.3425, + "eval_scitail-pairs-pos_samples_per_second": 85.379, + "eval_scitail-pairs-pos_steps_per_second": 2.134, + "step": 2874 + }, + { + "epoch": 0.45075282308657466, + "eval_xsum-pairs_loss": 1.0554753541946411, + "eval_xsum-pairs_runtime": 1.044, + "eval_xsum-pairs_samples_per_second": 191.573, + "eval_xsum-pairs_steps_per_second": 4.789, + "step": 2874 + }, + { + "epoch": 0.45075282308657466, + "eval_compression-pairs_loss": 0.6035106778144836, + "eval_compression-pairs_runtime": 0.241, + "eval_compression-pairs_samples_per_second": 830.038, + "eval_compression-pairs_steps_per_second": 20.751, + "step": 2874 + }, + { + "epoch": 0.45075282308657466, + "eval_sciq_pairs_loss": 8.811105728149414, + "eval_sciq_pairs_runtime": 9.0357, + "eval_sciq_pairs_samples_per_second": 22.134, + "eval_sciq_pairs_steps_per_second": 0.553, + "step": 2874 + }, + { + "epoch": 0.45075282308657466, + "eval_qasc_pairs_loss": 1.615903377532959, + "eval_qasc_pairs_runtime": 1.214, + "eval_qasc_pairs_samples_per_second": 164.746, + "eval_qasc_pairs_steps_per_second": 4.119, + "step": 2874 + }, + { + "epoch": 0.45075282308657466, + "eval_openbookqa_pairs_loss": 3.4049320220947266, + "eval_openbookqa_pairs_runtime": 1.0554, + "eval_openbookqa_pairs_samples_per_second": 189.509, + "eval_openbookqa_pairs_steps_per_second": 4.738, + "step": 2874 + }, + { + "epoch": 0.45075282308657466, + "eval_msmarco_pairs_loss": 2.3909060955047607, + "eval_msmarco_pairs_runtime": 2.5301, + "eval_msmarco_pairs_samples_per_second": 79.048, + "eval_msmarco_pairs_steps_per_second": 1.976, + "step": 2874 + }, + { + "epoch": 0.45075282308657466, + "eval_nq_pairs_loss": 2.794445753097534, + "eval_nq_pairs_runtime": 5.6752, + "eval_nq_pairs_samples_per_second": 35.241, + "eval_nq_pairs_steps_per_second": 0.881, + "step": 2874 + }, + { + "epoch": 0.45075282308657466, + "eval_trivia_pairs_loss": 2.753361701965332, + "eval_trivia_pairs_runtime": 9.0766, + "eval_trivia_pairs_samples_per_second": 22.035, + "eval_trivia_pairs_steps_per_second": 0.551, + "step": 2874 + }, + { + "epoch": 0.45075282308657466, + "eval_quora_pairs_loss": 0.205492302775383, + "eval_quora_pairs_runtime": 0.6182, + "eval_quora_pairs_samples_per_second": 323.536, + "eval_quora_pairs_steps_per_second": 8.088, + "step": 2874 + }, + { + "epoch": 0.45075282308657466, + "eval_gooaq_pairs_loss": 2.038878917694092, + "eval_gooaq_pairs_runtime": 1.5488, + "eval_gooaq_pairs_samples_per_second": 129.134, + "eval_gooaq_pairs_steps_per_second": 3.228, + "step": 2874 + }, + { + "epoch": 0.45075282308657466, + "eval_mrpc_pairs_loss": 0.3230588436126709, + "eval_mrpc_pairs_runtime": 0.2358, + "eval_mrpc_pairs_samples_per_second": 848.229, + "eval_mrpc_pairs_steps_per_second": 21.206, + "step": 2874 + }, + { + "epoch": 0.451693851944793, + "grad_norm": 5.271574020385742, + "learning_rate": 2.103513174404015e-05, + "loss": 2.6918, + "step": 2880 + }, + { + "epoch": 0.4592220828105395, + "grad_norm": 21.954103469848633, + "learning_rate": 2.1386449184441654e-05, + "loss": 2.8354, + "step": 2928 + }, + { + "epoch": 0.46675031367628605, + "grad_norm": 28.671293258666992, + "learning_rate": 2.173776662484316e-05, + "loss": 2.9499, + "step": 2976 + }, + { + "epoch": 0.4742785445420326, + "grad_norm": 26.562397003173828, + "learning_rate": 2.2089084065244666e-05, + "loss": 2.6211, + "step": 3024 + }, + { + "epoch": 0.48180677540777916, + "grad_norm": 48.511756896972656, + "learning_rate": 2.2440401505646173e-05, + "loss": 3.3356, + "step": 3072 + }, + { + "epoch": 0.4893350062735257, + "grad_norm": 46.71563720703125, + "learning_rate": 2.2791718946047677e-05, + "loss": 2.846, + "step": 3120 + }, + { + "epoch": 0.4968632371392723, + "grad_norm": 24.524322509765625, + "learning_rate": 2.3143036386449184e-05, + "loss": 2.4866, + "step": 3168 + }, + { + "epoch": 0.5043914680050188, + "grad_norm": 85.22843933105469, + "learning_rate": 2.3494353826850688e-05, + "loss": 2.6334, + "step": 3216 + }, + { + "epoch": 0.5119196988707654, + "grad_norm": 28.435443878173828, + "learning_rate": 2.384567126725219e-05, + "loss": 2.6118, + "step": 3264 + }, + { + "epoch": 0.5194479297365119, + "grad_norm": 21.590103149414062, + "learning_rate": 2.41969887076537e-05, + "loss": 2.5833, + "step": 3312 + }, + { + "epoch": 0.5258782936010038, + "eval_nli-pairs_loss": 2.0752949714660645, + "eval_nli-pairs_runtime": 4.0304, + "eval_nli-pairs_samples_per_second": 49.623, + "eval_nli-pairs_steps_per_second": 1.241, + "eval_sts-test_pearson_cosine": 0.7401847199967786, + "eval_sts-test_pearson_dot": 0.5441501995975192, + "eval_sts-test_pearson_euclidean": 0.7344996320188322, + "eval_sts-test_pearson_manhattan": 0.7394640598472787, + "eval_sts-test_pearson_max": 0.7401847199967786, + "eval_sts-test_spearman_cosine": 0.7300085598018916, + "eval_sts-test_spearman_dot": 0.5241747185593542, + "eval_sts-test_spearman_euclidean": 0.7194131601167465, + "eval_sts-test_spearman_manhattan": 0.726961581928453, + "eval_sts-test_spearman_max": 0.7300085598018916, + "step": 3353 + }, + { + "epoch": 0.5258782936010038, + "eval_vitaminc-pairs_loss": 6.441956996917725, + "eval_vitaminc-pairs_runtime": 1.4416, + "eval_vitaminc-pairs_samples_per_second": 115.149, + "eval_vitaminc-pairs_steps_per_second": 2.775, + "step": 3353 + }, + { + "epoch": 0.5258782936010038, + "eval_sts-label_loss": 4.200085639953613, + "eval_sts-label_runtime": 0.3949, + "eval_sts-label_samples_per_second": 506.436, + "eval_sts-label_steps_per_second": 12.661, + "step": 3353 + }, + { + "epoch": 0.5258782936010038, + "eval_qnli-contrastive_loss": 0.5195684432983398, + "eval_qnli-contrastive_runtime": 0.2809, + "eval_qnli-contrastive_samples_per_second": 712.107, + "eval_qnli-contrastive_steps_per_second": 17.803, + "step": 3353 + }, + { + "epoch": 0.5258782936010038, + "eval_scitail-pairs-qa_loss": 0.35189124941825867, + "eval_scitail-pairs-qa_runtime": 1.0578, + "eval_scitail-pairs-qa_samples_per_second": 189.064, + "eval_scitail-pairs-qa_steps_per_second": 4.727, + "step": 3353 + }, + { + "epoch": 0.5258782936010038, + "eval_scitail-pairs-pos_loss": 0.8873756527900696, + "eval_scitail-pairs-pos_runtime": 2.4029, + "eval_scitail-pairs-pos_samples_per_second": 83.232, + "eval_scitail-pairs-pos_steps_per_second": 2.081, + "step": 3353 + }, + { + "epoch": 0.5258782936010038, + "eval_xsum-pairs_loss": 0.939339280128479, + "eval_xsum-pairs_runtime": 1.041, + "eval_xsum-pairs_samples_per_second": 192.121, + "eval_xsum-pairs_steps_per_second": 4.803, + "step": 3353 + }, + { + "epoch": 0.5258782936010038, + "eval_compression-pairs_loss": 0.5007131695747375, + "eval_compression-pairs_runtime": 0.2338, + "eval_compression-pairs_samples_per_second": 855.479, + "eval_compression-pairs_steps_per_second": 21.387, + "step": 3353 + }, + { + "epoch": 0.5258782936010038, + "eval_sciq_pairs_loss": 8.558987617492676, + "eval_sciq_pairs_runtime": 9.0984, + "eval_sciq_pairs_samples_per_second": 21.982, + "eval_sciq_pairs_steps_per_second": 0.55, + "step": 3353 + }, + { + "epoch": 0.5258782936010038, + "eval_qasc_pairs_loss": 1.4318852424621582, + "eval_qasc_pairs_runtime": 1.2286, + "eval_qasc_pairs_samples_per_second": 162.79, + "eval_qasc_pairs_steps_per_second": 4.07, + "step": 3353 + }, + { + "epoch": 0.5258782936010038, + "eval_openbookqa_pairs_loss": 3.1973114013671875, + "eval_openbookqa_pairs_runtime": 1.0491, + "eval_openbookqa_pairs_samples_per_second": 190.633, + "eval_openbookqa_pairs_steps_per_second": 4.766, + "step": 3353 + }, + { + "epoch": 0.5258782936010038, + "eval_msmarco_pairs_loss": 2.2080254554748535, + "eval_msmarco_pairs_runtime": 2.5223, + "eval_msmarco_pairs_samples_per_second": 79.294, + "eval_msmarco_pairs_steps_per_second": 1.982, + "step": 3353 + }, + { + "epoch": 0.5258782936010038, + "eval_nq_pairs_loss": 2.5810558795928955, + "eval_nq_pairs_runtime": 5.6341, + "eval_nq_pairs_samples_per_second": 35.498, + "eval_nq_pairs_steps_per_second": 0.887, + "step": 3353 + }, + { + "epoch": 0.5258782936010038, + "eval_trivia_pairs_loss": 2.655771255493164, + "eval_trivia_pairs_runtime": 9.0716, + "eval_trivia_pairs_samples_per_second": 22.047, + "eval_trivia_pairs_steps_per_second": 0.551, + "step": 3353 + }, + { + "epoch": 0.5258782936010038, + "eval_quora_pairs_loss": 0.5028819441795349, + "eval_quora_pairs_runtime": 0.6144, + "eval_quora_pairs_samples_per_second": 325.522, + "eval_quora_pairs_steps_per_second": 8.138, + "step": 3353 + }, + { + "epoch": 0.5258782936010038, + "eval_gooaq_pairs_loss": 1.8867437839508057, + "eval_gooaq_pairs_runtime": 1.5505, + "eval_gooaq_pairs_samples_per_second": 128.994, + "eval_gooaq_pairs_steps_per_second": 3.225, + "step": 3353 + }, + { + "epoch": 0.5258782936010038, + "eval_mrpc_pairs_loss": 0.2580638825893402, + "eval_mrpc_pairs_runtime": 0.2364, + "eval_mrpc_pairs_samples_per_second": 846.008, + "eval_mrpc_pairs_steps_per_second": 21.15, + "step": 3353 + }, + { + "epoch": 0.5269761606022585, + "grad_norm": 2.6962711811065674, + "learning_rate": 2.4548306148055206e-05, + "loss": 2.3251, + "step": 3360 + }, + { + "epoch": 0.534504391468005, + "grad_norm": 35.47948455810547, + "learning_rate": 2.489962358845671e-05, + "loss": 2.8494, + "step": 3408 + }, + { + "epoch": 0.5420326223337516, + "grad_norm": 8.13453483581543, + "learning_rate": 2.5250941028858214e-05, + "loss": 2.4009, + "step": 3456 + }, + { + "epoch": 0.5495608531994981, + "grad_norm": 20.041057586669922, + "learning_rate": 2.560225846925972e-05, + "loss": 2.5952, + "step": 3504 + }, + { + "epoch": 0.5570890840652447, + "grad_norm": 23.942073822021484, + "learning_rate": 2.595357590966123e-05, + "loss": 2.2798, + "step": 3552 + }, + { + "epoch": 0.5646173149309912, + "grad_norm": 17.675006866455078, + "learning_rate": 2.6304893350062732e-05, + "loss": 2.308, + "step": 3600 + }, + { + "epoch": 0.5721455457967378, + "grad_norm": 24.20000457763672, + "learning_rate": 2.6656210790464236e-05, + "loss": 2.122, + "step": 3648 + }, + { + "epoch": 0.5796737766624843, + "grad_norm": 30.06256866455078, + "learning_rate": 2.7007528230865747e-05, + "loss": 2.7901, + "step": 3696 + }, + { + "epoch": 0.5872020075282308, + "grad_norm": 22.547115325927734, + "learning_rate": 2.735884567126725e-05, + "loss": 2.0671, + "step": 3744 + }, + { + "epoch": 0.5947302383939774, + "grad_norm": 34.11716079711914, + "learning_rate": 2.7710163111668754e-05, + "loss": 2.366, + "step": 3792 + }, + { + "epoch": 0.6010037641154329, + "eval_nli-pairs_loss": 1.8900150060653687, + "eval_nli-pairs_runtime": 4.0481, + "eval_nli-pairs_samples_per_second": 49.406, + "eval_nli-pairs_steps_per_second": 1.235, + "eval_sts-test_pearson_cosine": 0.752143976340549, + "eval_sts-test_pearson_dot": 0.5694102087200895, + "eval_sts-test_pearson_euclidean": 0.7457585181878474, + "eval_sts-test_pearson_manhattan": 0.7525316002813096, + "eval_sts-test_pearson_max": 0.7525316002813096, + "eval_sts-test_spearman_cosine": 0.7404216272264129, + "eval_sts-test_spearman_dot": 0.5485789739808921, + "eval_sts-test_spearman_euclidean": 0.728675089641457, + "eval_sts-test_spearman_manhattan": 0.7367562035227414, + "eval_sts-test_spearman_max": 0.7404216272264129, + "step": 3832 + }, + { + "epoch": 0.6010037641154329, + "eval_vitaminc-pairs_loss": 6.0831098556518555, + "eval_vitaminc-pairs_runtime": 1.4528, + "eval_vitaminc-pairs_samples_per_second": 114.264, + "eval_vitaminc-pairs_steps_per_second": 2.753, + "step": 3832 + }, + { + "epoch": 0.6010037641154329, + "eval_sts-label_loss": 4.197264671325684, + "eval_sts-label_runtime": 0.4176, + "eval_sts-label_samples_per_second": 478.893, + "eval_sts-label_steps_per_second": 11.972, + "step": 3832 + }, + { + "epoch": 0.6010037641154329, + "eval_qnli-contrastive_loss": 0.5115653872489929, + "eval_qnli-contrastive_runtime": 0.3027, + "eval_qnli-contrastive_samples_per_second": 660.784, + "eval_qnli-contrastive_steps_per_second": 16.52, + "step": 3832 + }, + { + "epoch": 0.6010037641154329, + "eval_scitail-pairs-qa_loss": 0.29788386821746826, + "eval_scitail-pairs-qa_runtime": 1.0654, + "eval_scitail-pairs-qa_samples_per_second": 187.719, + "eval_scitail-pairs-qa_steps_per_second": 4.693, + "step": 3832 + }, + { + "epoch": 0.6010037641154329, + "eval_scitail-pairs-pos_loss": 0.8727617859840393, + "eval_scitail-pairs-pos_runtime": 2.3677, + "eval_scitail-pairs-pos_samples_per_second": 84.471, + "eval_scitail-pairs-pos_steps_per_second": 2.112, + "step": 3832 + }, + { + "epoch": 0.6010037641154329, + "eval_xsum-pairs_loss": 0.8608022928237915, + "eval_xsum-pairs_runtime": 1.0435, + "eval_xsum-pairs_samples_per_second": 191.671, + "eval_xsum-pairs_steps_per_second": 4.792, + "step": 3832 + }, + { + "epoch": 0.6010037641154329, + "eval_compression-pairs_loss": 0.4411359429359436, + "eval_compression-pairs_runtime": 0.2345, + "eval_compression-pairs_samples_per_second": 852.821, + "eval_compression-pairs_steps_per_second": 21.321, + "step": 3832 + }, + { + "epoch": 0.6010037641154329, + "eval_sciq_pairs_loss": 8.294719696044922, + "eval_sciq_pairs_runtime": 9.2141, + "eval_sciq_pairs_samples_per_second": 21.706, + "eval_sciq_pairs_steps_per_second": 0.543, + "step": 3832 + }, + { + "epoch": 0.6010037641154329, + "eval_qasc_pairs_loss": 1.1894803047180176, + "eval_qasc_pairs_runtime": 1.2518, + "eval_qasc_pairs_samples_per_second": 159.774, + "eval_qasc_pairs_steps_per_second": 3.994, + "step": 3832 + }, + { + "epoch": 0.6010037641154329, + "eval_openbookqa_pairs_loss": 2.8579885959625244, + "eval_openbookqa_pairs_runtime": 1.0874, + "eval_openbookqa_pairs_samples_per_second": 183.92, + "eval_openbookqa_pairs_steps_per_second": 4.598, + "step": 3832 + }, + { + "epoch": 0.6010037641154329, + "eval_msmarco_pairs_loss": 1.9733755588531494, + "eval_msmarco_pairs_runtime": 2.5486, + "eval_msmarco_pairs_samples_per_second": 78.476, + "eval_msmarco_pairs_steps_per_second": 1.962, + "step": 3832 + }, + { + "epoch": 0.6010037641154329, + "eval_nq_pairs_loss": 2.206907033920288, + "eval_nq_pairs_runtime": 5.7528, + "eval_nq_pairs_samples_per_second": 34.766, + "eval_nq_pairs_steps_per_second": 0.869, + "step": 3832 + }, + { + "epoch": 0.6010037641154329, + "eval_trivia_pairs_loss": 2.332620620727539, + "eval_trivia_pairs_runtime": 9.1703, + "eval_trivia_pairs_samples_per_second": 21.809, + "eval_trivia_pairs_steps_per_second": 0.545, + "step": 3832 + }, + { + "epoch": 0.6010037641154329, + "eval_quora_pairs_loss": 0.48870089650154114, + "eval_quora_pairs_runtime": 0.6491, + "eval_quora_pairs_samples_per_second": 308.142, + "eval_quora_pairs_steps_per_second": 7.704, + "step": 3832 + }, + { + "epoch": 0.6010037641154329, + "eval_gooaq_pairs_loss": 1.598087191581726, + "eval_gooaq_pairs_runtime": 1.5759, + "eval_gooaq_pairs_samples_per_second": 126.912, + "eval_gooaq_pairs_steps_per_second": 3.173, + "step": 3832 + }, + { + "epoch": 0.6010037641154329, + "eval_mrpc_pairs_loss": 0.2343733161687851, + "eval_mrpc_pairs_runtime": 0.2484, + "eval_mrpc_pairs_samples_per_second": 805.097, + "eval_mrpc_pairs_steps_per_second": 20.127, + "step": 3832 + }, + { + "epoch": 0.6022584692597239, + "grad_norm": 1.486786127090454, + "learning_rate": 2.806148055207026e-05, + "loss": 1.9614, + "step": 3840 + }, + { + "epoch": 0.6097867001254705, + "grad_norm": 23.297300338745117, + "learning_rate": 2.841279799247177e-05, + "loss": 2.3589, + "step": 3888 + }, + { + "epoch": 0.617314930991217, + "grad_norm": 16.00516700744629, + "learning_rate": 2.8764115432873273e-05, + "loss": 2.1475, + "step": 3936 + }, + { + "epoch": 0.6248431618569636, + "grad_norm": 24.357616424560547, + "learning_rate": 2.9115432873274777e-05, + "loss": 2.1312, + "step": 3984 + }, + { + "epoch": 0.6323713927227101, + "grad_norm": 28.798917770385742, + "learning_rate": 2.946675031367628e-05, + "loss": 2.5716, + "step": 4032 + }, + { + "epoch": 0.6398996235884568, + "grad_norm": 18.239490509033203, + "learning_rate": 2.981806775407779e-05, + "loss": 2.2249, + "step": 4080 + }, + { + "epoch": 0.6474278544542033, + "grad_norm": 19.50409507751465, + "learning_rate": 3.0169385194479295e-05, + "loss": 2.6331, + "step": 4128 + }, + { + "epoch": 0.6549560853199499, + "grad_norm": 12.110575675964355, + "learning_rate": 3.05207026348808e-05, + "loss": 2.7637, + "step": 4176 + }, + { + "epoch": 0.6624843161856964, + "grad_norm": 6.904999256134033, + "learning_rate": 3.087202007528231e-05, + "loss": 1.8973, + "step": 4224 + }, + { + "epoch": 0.6700125470514429, + "grad_norm": 9.007365226745605, + "learning_rate": 3.1223337515683813e-05, + "loss": 2.3181, + "step": 4272 + }, + { + "epoch": 0.676129234629862, + "eval_nli-pairs_loss": 1.7111084461212158, + "eval_nli-pairs_runtime": 4.0305, + "eval_nli-pairs_samples_per_second": 49.622, + "eval_nli-pairs_steps_per_second": 1.241, + "eval_sts-test_pearson_cosine": 0.7375865838793885, + "eval_sts-test_pearson_dot": 0.5355907015359193, + "eval_sts-test_pearson_euclidean": 0.7266850031847317, + "eval_sts-test_pearson_manhattan": 0.7357621558005936, + "eval_sts-test_pearson_max": 0.7375865838793885, + "eval_sts-test_spearman_cosine": 0.7273524041973777, + "eval_sts-test_spearman_dot": 0.5084902224306463, + "eval_sts-test_spearman_euclidean": 0.7071419579928555, + "eval_sts-test_spearman_manhattan": 0.7177664681655631, + "eval_sts-test_spearman_max": 0.7273524041973777, + "step": 4311 + }, + { + "epoch": 0.676129234629862, + "eval_vitaminc-pairs_loss": 6.216845989227295, + "eval_vitaminc-pairs_runtime": 1.4703, + "eval_vitaminc-pairs_samples_per_second": 112.902, + "eval_vitaminc-pairs_steps_per_second": 2.721, + "step": 4311 + }, + { + "epoch": 0.676129234629862, + "eval_sts-label_loss": 4.384557723999023, + "eval_sts-label_runtime": 0.3912, + "eval_sts-label_samples_per_second": 511.22, + "eval_sts-label_steps_per_second": 12.78, + "step": 4311 + }, + { + "epoch": 0.676129234629862, + "eval_qnli-contrastive_loss": 0.40437957644462585, + "eval_qnli-contrastive_runtime": 0.281, + "eval_qnli-contrastive_samples_per_second": 711.864, + "eval_qnli-contrastive_steps_per_second": 17.797, + "step": 4311 + }, + { + "epoch": 0.676129234629862, + "eval_scitail-pairs-qa_loss": 0.2210184931755066, + "eval_scitail-pairs-qa_runtime": 1.0575, + "eval_scitail-pairs-qa_samples_per_second": 189.117, + "eval_scitail-pairs-qa_steps_per_second": 4.728, + "step": 4311 + }, + { + "epoch": 0.676129234629862, + "eval_scitail-pairs-pos_loss": 0.9065079689025879, + "eval_scitail-pairs-pos_runtime": 2.3488, + "eval_scitail-pairs-pos_samples_per_second": 85.151, + "eval_scitail-pairs-pos_steps_per_second": 2.129, + "step": 4311 + }, + { + "epoch": 0.676129234629862, + "eval_xsum-pairs_loss": 0.8169436454772949, + "eval_xsum-pairs_runtime": 1.0409, + "eval_xsum-pairs_samples_per_second": 192.145, + "eval_xsum-pairs_steps_per_second": 4.804, + "step": 4311 + }, + { + "epoch": 0.676129234629862, + "eval_compression-pairs_loss": 0.391815721988678, + "eval_compression-pairs_runtime": 0.2361, + "eval_compression-pairs_samples_per_second": 847.066, + "eval_compression-pairs_steps_per_second": 21.177, + "step": 4311 + }, + { + "epoch": 0.676129234629862, + "eval_sciq_pairs_loss": 0.6230970025062561, + "eval_sciq_pairs_runtime": 9.0874, + "eval_sciq_pairs_samples_per_second": 22.008, + "eval_sciq_pairs_steps_per_second": 0.55, + "step": 4311 + }, + { + "epoch": 0.676129234629862, + "eval_qasc_pairs_loss": 1.1559942960739136, + "eval_qasc_pairs_runtime": 1.2507, + "eval_qasc_pairs_samples_per_second": 159.907, + "eval_qasc_pairs_steps_per_second": 3.998, + "step": 4311 + }, + { + "epoch": 0.676129234629862, + "eval_openbookqa_pairs_loss": 2.8303356170654297, + "eval_openbookqa_pairs_runtime": 1.0524, + "eval_openbookqa_pairs_samples_per_second": 190.041, + "eval_openbookqa_pairs_steps_per_second": 4.751, + "step": 4311 + }, + { + "epoch": 0.676129234629862, + "eval_msmarco_pairs_loss": 1.9672399759292603, + "eval_msmarco_pairs_runtime": 2.518, + "eval_msmarco_pairs_samples_per_second": 79.428, + "eval_msmarco_pairs_steps_per_second": 1.986, + "step": 4311 + }, + { + "epoch": 0.676129234629862, + "eval_nq_pairs_loss": 2.169950008392334, + "eval_nq_pairs_runtime": 5.6541, + "eval_nq_pairs_samples_per_second": 35.372, + "eval_nq_pairs_steps_per_second": 0.884, + "step": 4311 + }, + { + "epoch": 0.676129234629862, + "eval_trivia_pairs_loss": 2.198312520980835, + "eval_trivia_pairs_runtime": 9.0535, + "eval_trivia_pairs_samples_per_second": 22.091, + "eval_trivia_pairs_steps_per_second": 0.552, + "step": 4311 + }, + { + "epoch": 0.676129234629862, + "eval_quora_pairs_loss": 0.3780948519706726, + "eval_quora_pairs_runtime": 0.6375, + "eval_quora_pairs_samples_per_second": 313.737, + "eval_quora_pairs_steps_per_second": 7.843, + "step": 4311 + }, + { + "epoch": 0.676129234629862, + "eval_gooaq_pairs_loss": 1.5646275281906128, + "eval_gooaq_pairs_runtime": 1.5531, + "eval_gooaq_pairs_samples_per_second": 128.772, + "eval_gooaq_pairs_steps_per_second": 3.219, + "step": 4311 + }, + { + "epoch": 0.676129234629862, + "eval_mrpc_pairs_loss": 0.18545588850975037, + "eval_mrpc_pairs_runtime": 0.2382, + "eval_mrpc_pairs_samples_per_second": 839.569, + "eval_mrpc_pairs_steps_per_second": 20.989, + "step": 4311 + }, + { + "epoch": 0.6775407779171895, + "grad_norm": 15.639892578125, + "learning_rate": 3.157465495608532e-05, + "loss": 2.2329, + "step": 4320 + }, + { + "epoch": 0.685069008782936, + "grad_norm": 7.185269832611084, + "learning_rate": 3.192597239648682e-05, + "loss": 2.7864, + "step": 4368 + }, + { + "epoch": 0.6925972396486826, + "grad_norm": 10.165898323059082, + "learning_rate": 3.227728983688833e-05, + "loss": 2.5277, + "step": 4416 + }, + { + "epoch": 0.7001254705144291, + "grad_norm": 20.132612228393555, + "learning_rate": 3.2628607277289836e-05, + "loss": 2.526, + "step": 4464 + }, + { + "epoch": 0.7076537013801757, + "grad_norm": 6.147126197814941, + "learning_rate": 3.297992471769134e-05, + "loss": 1.5993, + "step": 4512 + }, + { + "epoch": 0.7151819322459222, + "grad_norm": 9.857342720031738, + "learning_rate": 3.3331242158092843e-05, + "loss": 2.0452, + "step": 4560 + }, + { + "epoch": 0.7227101631116688, + "grad_norm": 22.434364318847656, + "learning_rate": 3.3682559598494354e-05, + "loss": 2.0458, + "step": 4608 + }, + { + "epoch": 0.7302383939774153, + "grad_norm": 22.420066833496094, + "learning_rate": 3.403387703889586e-05, + "loss": 1.854, + "step": 4656 + }, + { + "epoch": 0.7377666248431619, + "grad_norm": 24.213205337524414, + "learning_rate": 3.438519447929736e-05, + "loss": 2.31, + "step": 4704 + }, + { + "epoch": 0.7452948557089084, + "grad_norm": 2.1184492111206055, + "learning_rate": 3.473651191969887e-05, + "loss": 1.7484, + "step": 4752 + }, + { + "epoch": 0.751254705144291, + "eval_nli-pairs_loss": 1.5204579830169678, + "eval_nli-pairs_runtime": 4.3227, + "eval_nli-pairs_samples_per_second": 46.267, + "eval_nli-pairs_steps_per_second": 1.157, + "eval_sts-test_pearson_cosine": 0.753550468294361, + "eval_sts-test_pearson_dot": 0.576164453162354, + "eval_sts-test_pearson_euclidean": 0.7433413992355353, + "eval_sts-test_pearson_manhattan": 0.7483177470711824, + "eval_sts-test_pearson_max": 0.753550468294361, + "eval_sts-test_spearman_cosine": 0.7510075785449373, + "eval_sts-test_spearman_dot": 0.5438417987754244, + "eval_sts-test_spearman_euclidean": 0.7271758422639625, + "eval_sts-test_spearman_manhattan": 0.7334567781451864, + "eval_sts-test_spearman_max": 0.7510075785449373, + "step": 4790 + }, + { + "epoch": 0.751254705144291, + "eval_vitaminc-pairs_loss": 5.992164134979248, + "eval_vitaminc-pairs_runtime": 1.4716, + "eval_vitaminc-pairs_samples_per_second": 112.802, + "eval_vitaminc-pairs_steps_per_second": 2.718, + "step": 4790 + }, + { + "epoch": 0.751254705144291, + "eval_sts-label_loss": 4.175446033477783, + "eval_sts-label_runtime": 0.3991, + "eval_sts-label_samples_per_second": 501.105, + "eval_sts-label_steps_per_second": 12.528, + "step": 4790 + }, + { + "epoch": 0.751254705144291, + "eval_qnli-contrastive_loss": 0.4543713629245758, + "eval_qnli-contrastive_runtime": 0.2765, + "eval_qnli-contrastive_samples_per_second": 723.231, + "eval_qnli-contrastive_steps_per_second": 18.081, + "step": 4790 + }, + { + "epoch": 0.751254705144291, + "eval_scitail-pairs-qa_loss": 0.20784999430179596, + "eval_scitail-pairs-qa_runtime": 1.0567, + "eval_scitail-pairs-qa_samples_per_second": 189.26, + "eval_scitail-pairs-qa_steps_per_second": 4.732, + "step": 4790 + }, + { + "epoch": 0.751254705144291, + "eval_scitail-pairs-pos_loss": 0.8679056167602539, + "eval_scitail-pairs-pos_runtime": 2.3456, + "eval_scitail-pairs-pos_samples_per_second": 85.268, + "eval_scitail-pairs-pos_steps_per_second": 2.132, + "step": 4790 + }, + { + "epoch": 0.751254705144291, + "eval_xsum-pairs_loss": 0.6944636702537537, + "eval_xsum-pairs_runtime": 1.0379, + "eval_xsum-pairs_samples_per_second": 192.694, + "eval_xsum-pairs_steps_per_second": 4.817, + "step": 4790 + }, + { + "epoch": 0.751254705144291, + "eval_compression-pairs_loss": 0.34194332361221313, + "eval_compression-pairs_runtime": 0.2354, + "eval_compression-pairs_samples_per_second": 849.798, + "eval_compression-pairs_steps_per_second": 21.245, + "step": 4790 + }, + { + "epoch": 0.751254705144291, + "eval_sciq_pairs_loss": 7.352969646453857, + "eval_sciq_pairs_runtime": 9.0502, + "eval_sciq_pairs_samples_per_second": 22.099, + "eval_sciq_pairs_steps_per_second": 0.552, + "step": 4790 + }, + { + "epoch": 0.751254705144291, + "eval_qasc_pairs_loss": 0.9425787329673767, + "eval_qasc_pairs_runtime": 1.2081, + "eval_qasc_pairs_samples_per_second": 165.543, + "eval_qasc_pairs_steps_per_second": 4.139, + "step": 4790 + }, + { + "epoch": 0.751254705144291, + "eval_openbookqa_pairs_loss": 2.5082011222839355, + "eval_openbookqa_pairs_runtime": 1.0492, + "eval_openbookqa_pairs_samples_per_second": 190.618, + "eval_openbookqa_pairs_steps_per_second": 4.765, + "step": 4790 + }, + { + "epoch": 0.751254705144291, + "eval_msmarco_pairs_loss": 1.696744441986084, + "eval_msmarco_pairs_runtime": 2.5162, + "eval_msmarco_pairs_samples_per_second": 79.484, + "eval_msmarco_pairs_steps_per_second": 1.987, + "step": 4790 + }, + { + "epoch": 0.751254705144291, + "eval_nq_pairs_loss": 1.8095602989196777, + "eval_nq_pairs_runtime": 5.6348, + "eval_nq_pairs_samples_per_second": 35.494, + "eval_nq_pairs_steps_per_second": 0.887, + "step": 4790 + }, + { + "epoch": 0.751254705144291, + "eval_trivia_pairs_loss": 2.0272486209869385, + "eval_trivia_pairs_runtime": 9.0671, + "eval_trivia_pairs_samples_per_second": 22.058, + "eval_trivia_pairs_steps_per_second": 0.551, + "step": 4790 + }, + { + "epoch": 0.751254705144291, + "eval_quora_pairs_loss": 0.15705542266368866, + "eval_quora_pairs_runtime": 0.6326, + "eval_quora_pairs_samples_per_second": 316.138, + "eval_quora_pairs_steps_per_second": 7.903, + "step": 4790 + }, + { + "epoch": 0.751254705144291, + "eval_gooaq_pairs_loss": 1.374332308769226, + "eval_gooaq_pairs_runtime": 1.5484, + "eval_gooaq_pairs_samples_per_second": 129.164, + "eval_gooaq_pairs_steps_per_second": 3.229, + "step": 4790 + }, + { + "epoch": 0.751254705144291, + "eval_mrpc_pairs_loss": 0.17204828560352325, + "eval_mrpc_pairs_runtime": 0.2358, + "eval_mrpc_pairs_samples_per_second": 848.332, + "eval_mrpc_pairs_steps_per_second": 21.208, + "step": 4790 + }, + { + "epoch": 0.7528230865746549, + "grad_norm": 1.4021190404891968, + "learning_rate": 3.4999758305020584e-05, + "loss": 1.7113, + "step": 4800 + }, + { + "epoch": 0.7603513174404015, + "grad_norm": 19.776817321777344, + "learning_rate": 3.499395795931671e-05, + "loss": 2.3696, + "step": 4848 + }, + { + "epoch": 0.767879548306148, + "grad_norm": 28.693845748901367, + "learning_rate": 3.49804263115427e-05, + "loss": 2.2947, + "step": 4896 + }, + { + "epoch": 0.7754077791718946, + "grad_norm": 1.3631008863449097, + "learning_rate": 3.495916934189221e-05, + "loss": 1.8841, + "step": 4944 + }, + { + "epoch": 0.7829360100376411, + "grad_norm": 40.640262603759766, + "learning_rate": 3.4930196444697477e-05, + "loss": 2.084, + "step": 4992 + }, + { + "epoch": 0.7904642409033877, + "grad_norm": 20.45759391784668, + "learning_rate": 3.489352042427762e-05, + "loss": 1.9297, + "step": 5040 + }, + { + "epoch": 0.7979924717691342, + "grad_norm": 24.276058197021484, + "learning_rate": 3.484915748927982e-05, + "loss": 2.0521, + "step": 5088 + }, + { + "epoch": 0.8055207026348808, + "grad_norm": 24.93791389465332, + "learning_rate": 3.4797127245516105e-05, + "loss": 1.7092, + "step": 5136 + }, + { + "epoch": 0.8130489335006273, + "grad_norm": 25.131153106689453, + "learning_rate": 3.4737452687298694e-05, + "loss": 1.7394, + "step": 5184 + }, + { + "epoch": 0.820577164366374, + "grad_norm": 3.779459238052368, + "learning_rate": 3.467016018727788e-05, + "loss": 2.567, + "step": 5232 + }, + { + "epoch": 0.8263801756587202, + "eval_nli-pairs_loss": 1.4584167003631592, + "eval_nli-pairs_runtime": 3.9955, + "eval_nli-pairs_samples_per_second": 50.056, + "eval_nli-pairs_steps_per_second": 1.251, + "eval_sts-test_pearson_cosine": 0.7484577894142428, + "eval_sts-test_pearson_dot": 0.5289676422936789, + "eval_sts-test_pearson_euclidean": 0.743677607180833, + "eval_sts-test_pearson_manhattan": 0.7474581577502462, + "eval_sts-test_pearson_max": 0.7484577894142428, + "eval_sts-test_spearman_cosine": 0.7507798204197761, + "eval_sts-test_spearman_dot": 0.5016451185199292, + "eval_sts-test_spearman_euclidean": 0.7307379850546868, + "eval_sts-test_spearman_manhattan": 0.7367432097081014, + "eval_sts-test_spearman_max": 0.7507798204197761, + "step": 5269 + }, + { + "epoch": 0.8263801756587202, + "eval_vitaminc-pairs_loss": 5.56383752822876, + "eval_vitaminc-pairs_runtime": 1.4339, + "eval_vitaminc-pairs_samples_per_second": 115.772, + "eval_vitaminc-pairs_steps_per_second": 2.79, + "step": 5269 + }, + { + "epoch": 0.8263801756587202, + "eval_sts-label_loss": 4.355674743652344, + "eval_sts-label_runtime": 0.401, + "eval_sts-label_samples_per_second": 498.723, + "eval_sts-label_steps_per_second": 12.468, + "step": 5269 + }, + { + "epoch": 0.8263801756587202, + "eval_qnli-contrastive_loss": 0.2912294566631317, + "eval_qnli-contrastive_runtime": 0.2801, + "eval_qnli-contrastive_samples_per_second": 714.044, + "eval_qnli-contrastive_steps_per_second": 17.851, + "step": 5269 + }, + { + "epoch": 0.8263801756587202, + "eval_scitail-pairs-qa_loss": 0.19145721197128296, + "eval_scitail-pairs-qa_runtime": 1.0728, + "eval_scitail-pairs-qa_samples_per_second": 186.429, + "eval_scitail-pairs-qa_steps_per_second": 4.661, + "step": 5269 + }, + { + "epoch": 0.8263801756587202, + "eval_scitail-pairs-pos_loss": 0.7433645725250244, + "eval_scitail-pairs-pos_runtime": 2.3785, + "eval_scitail-pairs-pos_samples_per_second": 84.086, + "eval_scitail-pairs-pos_steps_per_second": 2.102, + "step": 5269 + }, + { + "epoch": 0.8263801756587202, + "eval_xsum-pairs_loss": 0.6145637631416321, + "eval_xsum-pairs_runtime": 1.0393, + "eval_xsum-pairs_samples_per_second": 192.428, + "eval_xsum-pairs_steps_per_second": 4.811, + "step": 5269 + }, + { + "epoch": 0.8263801756587202, + "eval_compression-pairs_loss": 0.29557526111602783, + "eval_compression-pairs_runtime": 0.2372, + "eval_compression-pairs_samples_per_second": 843.266, + "eval_compression-pairs_steps_per_second": 21.082, + "step": 5269 + }, + { + "epoch": 0.8263801756587202, + "eval_sciq_pairs_loss": 0.5607883930206299, + "eval_sciq_pairs_runtime": 9.0381, + "eval_sciq_pairs_samples_per_second": 22.128, + "eval_sciq_pairs_steps_per_second": 0.553, + "step": 5269 + }, + { + "epoch": 0.8263801756587202, + "eval_qasc_pairs_loss": 0.8776007294654846, + "eval_qasc_pairs_runtime": 1.2111, + "eval_qasc_pairs_samples_per_second": 165.141, + "eval_qasc_pairs_steps_per_second": 4.129, + "step": 5269 + }, + { + "epoch": 0.8263801756587202, + "eval_openbookqa_pairs_loss": 2.414658784866333, + "eval_openbookqa_pairs_runtime": 1.052, + "eval_openbookqa_pairs_samples_per_second": 190.106, + "eval_openbookqa_pairs_steps_per_second": 4.753, + "step": 5269 + }, + { + "epoch": 0.8263801756587202, + "eval_msmarco_pairs_loss": 1.615893840789795, + "eval_msmarco_pairs_runtime": 2.5183, + "eval_msmarco_pairs_samples_per_second": 79.419, + "eval_msmarco_pairs_steps_per_second": 1.985, + "step": 5269 + }, + { + "epoch": 0.8263801756587202, + "eval_nq_pairs_loss": 1.5788501501083374, + "eval_nq_pairs_runtime": 5.6429, + "eval_nq_pairs_samples_per_second": 35.443, + "eval_nq_pairs_steps_per_second": 0.886, + "step": 5269 + }, + { + "epoch": 0.8263801756587202, + "eval_trivia_pairs_loss": 1.8346160650253296, + "eval_trivia_pairs_runtime": 9.0522, + "eval_trivia_pairs_samples_per_second": 22.094, + "eval_trivia_pairs_steps_per_second": 0.552, + "step": 5269 + }, + { + "epoch": 0.8263801756587202, + "eval_quora_pairs_loss": 0.23968417942523956, + "eval_quora_pairs_runtime": 0.6179, + "eval_quora_pairs_samples_per_second": 323.693, + "eval_quora_pairs_steps_per_second": 8.092, + "step": 5269 + }, + { + "epoch": 0.8263801756587202, + "eval_gooaq_pairs_loss": 1.3264899253845215, + "eval_gooaq_pairs_runtime": 1.549, + "eval_gooaq_pairs_samples_per_second": 129.112, + "eval_gooaq_pairs_steps_per_second": 3.228, + "step": 5269 + }, + { + "epoch": 0.8263801756587202, + "eval_mrpc_pairs_loss": 0.14705294370651245, + "eval_mrpc_pairs_runtime": 0.2494, + "eval_mrpc_pairs_samples_per_second": 801.787, + "eval_mrpc_pairs_steps_per_second": 20.045, + "step": 5269 + }, + { + "epoch": 0.8281053952321205, + "grad_norm": 17.636714935302734, + "learning_rate": 3.459527948478686e-05, + "loss": 2.3021, + "step": 5280 + }, + { + "epoch": 0.835633626097867, + "grad_norm": 23.402650833129883, + "learning_rate": 3.4512843672698696e-05, + "loss": 1.5502, + "step": 5328 + }, + { + "epoch": 0.8431618569636136, + "grad_norm": 13.210539817810059, + "learning_rate": 3.4422889182801225e-05, + "loss": 1.7324, + "step": 5376 + }, + { + "epoch": 0.8506900878293601, + "grad_norm": 14.795612335205078, + "learning_rate": 3.4325455769696324e-05, + "loss": 1.8119, + "step": 5424 + }, + { + "epoch": 0.8582183186951067, + "grad_norm": 14.047534942626953, + "learning_rate": 3.422058649323072e-05, + "loss": 1.8507, + "step": 5472 + }, + { + "epoch": 0.8657465495608532, + "grad_norm": 0.7366377711296082, + "learning_rate": 3.4108327699466066e-05, + "loss": 1.7362, + "step": 5520 + }, + { + "epoch": 0.8732747804265998, + "grad_norm": 16.555519104003906, + "learning_rate": 3.398872900019673e-05, + "loss": 2.082, + "step": 5568 + }, + { + "epoch": 0.8808030112923463, + "grad_norm": 16.52071189880371, + "learning_rate": 3.386184325102423e-05, + "loss": 2.1483, + "step": 5616 + }, + { + "epoch": 0.8883312421580929, + "grad_norm": 16.03848648071289, + "learning_rate": 3.372772652799824e-05, + "loss": 1.3961, + "step": 5664 + }, + { + "epoch": 0.8958594730238394, + "grad_norm": 15.494946479797363, + "learning_rate": 3.358643810283421e-05, + "loss": 1.6331, + "step": 5712 + }, + { + "epoch": 0.9015056461731493, + "eval_nli-pairs_loss": 1.4454108476638794, + "eval_nli-pairs_runtime": 4.0041, + "eval_nli-pairs_samples_per_second": 49.949, + "eval_nli-pairs_steps_per_second": 1.249, + "eval_sts-test_pearson_cosine": 0.7644735043371528, + "eval_sts-test_pearson_dot": 0.5461512421131185, + "eval_sts-test_pearson_euclidean": 0.7520132891230207, + "eval_sts-test_pearson_manhattan": 0.7535418655995262, + "eval_sts-test_pearson_max": 0.7644735043371528, + "eval_sts-test_spearman_cosine": 0.7630644782411757, + "eval_sts-test_spearman_dot": 0.5239487411838791, + "eval_sts-test_spearman_euclidean": 0.7392793315112096, + "eval_sts-test_spearman_manhattan": 0.7426354353655322, + "eval_sts-test_spearman_max": 0.7630644782411757, + "step": 5748 + }, + { + "epoch": 0.9015056461731493, + "eval_vitaminc-pairs_loss": 5.328937530517578, + "eval_vitaminc-pairs_runtime": 1.4411, + "eval_vitaminc-pairs_samples_per_second": 115.19, + "eval_vitaminc-pairs_steps_per_second": 2.776, + "step": 5748 + }, + { + "epoch": 0.9015056461731493, + "eval_sts-label_loss": 4.186919212341309, + "eval_sts-label_runtime": 0.4046, + "eval_sts-label_samples_per_second": 494.357, + "eval_sts-label_steps_per_second": 12.359, + "step": 5748 + }, + { + "epoch": 0.9015056461731493, + "eval_qnli-contrastive_loss": 0.35569697618484497, + "eval_qnli-contrastive_runtime": 0.2807, + "eval_qnli-contrastive_samples_per_second": 712.531, + "eval_qnli-contrastive_steps_per_second": 17.813, + "step": 5748 + }, + { + "epoch": 0.9015056461731493, + "eval_scitail-pairs-qa_loss": 0.18789875507354736, + "eval_scitail-pairs-qa_runtime": 1.1135, + "eval_scitail-pairs-qa_samples_per_second": 179.622, + "eval_scitail-pairs-qa_steps_per_second": 4.491, + "step": 5748 + }, + { + "epoch": 0.9015056461731493, + "eval_scitail-pairs-pos_loss": 0.715129017829895, + "eval_scitail-pairs-pos_runtime": 2.3885, + "eval_scitail-pairs-pos_samples_per_second": 83.736, + "eval_scitail-pairs-pos_steps_per_second": 2.093, + "step": 5748 + }, + { + "epoch": 0.9015056461731493, + "eval_xsum-pairs_loss": 0.624691367149353, + "eval_xsum-pairs_runtime": 1.0408, + "eval_xsum-pairs_samples_per_second": 192.169, + "eval_xsum-pairs_steps_per_second": 4.804, + "step": 5748 + }, + { + "epoch": 0.9015056461731493, + "eval_compression-pairs_loss": 0.30371707677841187, + "eval_compression-pairs_runtime": 0.2581, + "eval_compression-pairs_samples_per_second": 774.8, + "eval_compression-pairs_steps_per_second": 19.37, + "step": 5748 + }, + { + "epoch": 0.9015056461731493, + "eval_sciq_pairs_loss": 0.5236299633979797, + "eval_sciq_pairs_runtime": 9.205, + "eval_sciq_pairs_samples_per_second": 21.727, + "eval_sciq_pairs_steps_per_second": 0.543, + "step": 5748 + }, + { + "epoch": 0.9015056461731493, + "eval_qasc_pairs_loss": 0.8543006777763367, + "eval_qasc_pairs_runtime": 1.238, + "eval_qasc_pairs_samples_per_second": 161.556, + "eval_qasc_pairs_steps_per_second": 4.039, + "step": 5748 + }, + { + "epoch": 0.9015056461731493, + "eval_openbookqa_pairs_loss": 2.3740031719207764, + "eval_openbookqa_pairs_runtime": 1.1145, + "eval_openbookqa_pairs_samples_per_second": 179.452, + "eval_openbookqa_pairs_steps_per_second": 4.486, + "step": 5748 + }, + { + "epoch": 0.9015056461731493, + "eval_msmarco_pairs_loss": 1.4328840970993042, + "eval_msmarco_pairs_runtime": 2.5725, + "eval_msmarco_pairs_samples_per_second": 77.746, + "eval_msmarco_pairs_steps_per_second": 1.944, + "step": 5748 + }, + { + "epoch": 0.9015056461731493, + "eval_nq_pairs_loss": 1.5343101024627686, + "eval_nq_pairs_runtime": 5.7416, + "eval_nq_pairs_samples_per_second": 34.833, + "eval_nq_pairs_steps_per_second": 0.871, + "step": 5748 + }, + { + "epoch": 0.9015056461731493, + "eval_trivia_pairs_loss": 1.7511711120605469, + "eval_trivia_pairs_runtime": 9.1035, + "eval_trivia_pairs_samples_per_second": 21.97, + "eval_trivia_pairs_steps_per_second": 0.549, + "step": 5748 + }, + { + "epoch": 0.9015056461731493, + "eval_quora_pairs_loss": 0.2826410233974457, + "eval_quora_pairs_runtime": 0.6444, + "eval_quora_pairs_samples_per_second": 310.354, + "eval_quora_pairs_steps_per_second": 7.759, + "step": 5748 + }, + { + "epoch": 0.9015056461731493, + "eval_gooaq_pairs_loss": 1.2128998041152954, + "eval_gooaq_pairs_runtime": 1.5919, + "eval_gooaq_pairs_samples_per_second": 125.634, + "eval_gooaq_pairs_steps_per_second": 3.141, + "step": 5748 + }, + { + "epoch": 0.9015056461731493, + "eval_mrpc_pairs_loss": 0.16468097269535065, + "eval_mrpc_pairs_runtime": 0.2496, + "eval_mrpc_pairs_samples_per_second": 801.385, + "eval_mrpc_pairs_steps_per_second": 20.035, + "step": 5748 + }, + { + "epoch": 0.903387703889586, + "grad_norm": 2.899136543273926, + "learning_rate": 3.3438040416718773e-05, + "loss": 1.9863, + "step": 5760 + }, + { + "epoch": 0.9109159347553325, + "grad_norm": 14.919694900512695, + "learning_rate": 3.3282599052714414e-05, + "loss": 1.6917, + "step": 5808 + }, + { + "epoch": 0.918444165621079, + "grad_norm": 1.626105785369873, + "learning_rate": 3.312018270677559e-05, + "loss": 1.7409, + "step": 5856 + }, + { + "epoch": 0.9259723964868256, + "grad_norm": 15.8577299118042, + "learning_rate": 3.295086315738918e-05, + "loss": 1.4397, + "step": 5904 + }, + { + "epoch": 0.9335006273525721, + "grad_norm": 10.255402565002441, + "learning_rate": 3.277471523385255e-05, + "loss": 1.2977, + "step": 5952 + }, + { + "epoch": 0.9410288582183187, + "grad_norm": 31.09028434753418, + "learning_rate": 3.259181678320349e-05, + "loss": 1.5429, + "step": 6000 + }, + { + "epoch": 0.9485570890840652, + "grad_norm": 15.244925498962402, + "learning_rate": 3.2402248635816294e-05, + "loss": 1.6939, + "step": 6048 + }, + { + "epoch": 0.9560853199498118, + "grad_norm": 19.404956817626953, + "learning_rate": 3.2206094569679564e-05, + "loss": 1.5573, + "step": 6096 + }, + { + "epoch": 0.9636135508155583, + "grad_norm": 16.607341766357422, + "learning_rate": 3.200344127337121e-05, + "loss": 1.5873, + "step": 6144 + }, + { + "epoch": 0.9711417816813049, + "grad_norm": 1.1432667970657349, + "learning_rate": 3.179437830774722e-05, + "loss": 1.8802, + "step": 6192 + }, + { + "epoch": 0.9766311166875784, + "eval_nli-pairs_loss": 1.3268091678619385, + "eval_nli-pairs_runtime": 4.0761, + "eval_nli-pairs_samples_per_second": 49.066, + "eval_nli-pairs_steps_per_second": 1.227, + "eval_sts-test_pearson_cosine": 0.7536875808596679, + "eval_sts-test_pearson_dot": 0.5240503602011544, + "eval_sts-test_pearson_euclidean": 0.7440772899082505, + "eval_sts-test_pearson_manhattan": 0.7465814532083983, + "eval_sts-test_pearson_max": 0.7536875808596679, + "eval_sts-test_spearman_cosine": 0.7584726286921011, + "eval_sts-test_spearman_dot": 0.49490205708473545, + "eval_sts-test_spearman_euclidean": 0.7307844981527315, + "eval_sts-test_spearman_manhattan": 0.7362283105144983, + "eval_sts-test_spearman_max": 0.7584726286921011, + "step": 6227 + }, + { + "epoch": 0.9766311166875784, + "eval_vitaminc-pairs_loss": 5.498671054840088, + "eval_vitaminc-pairs_runtime": 1.4379, + "eval_vitaminc-pairs_samples_per_second": 115.449, + "eval_vitaminc-pairs_steps_per_second": 2.782, + "step": 6227 + }, + { + "epoch": 0.9766311166875784, + "eval_sts-label_loss": 4.3516716957092285, + "eval_sts-label_runtime": 0.3976, + "eval_sts-label_samples_per_second": 503.027, + "eval_sts-label_steps_per_second": 12.576, + "step": 6227 + }, + { + "epoch": 0.9766311166875784, + "eval_qnli-contrastive_loss": 0.30157506465911865, + "eval_qnli-contrastive_runtime": 0.2787, + "eval_qnli-contrastive_samples_per_second": 717.742, + "eval_qnli-contrastive_steps_per_second": 17.944, + "step": 6227 + }, + { + "epoch": 0.9766311166875784, + "eval_scitail-pairs-qa_loss": 0.14389516413211823, + "eval_scitail-pairs-qa_runtime": 1.0465, + "eval_scitail-pairs-qa_samples_per_second": 191.121, + "eval_scitail-pairs-qa_steps_per_second": 4.778, + "step": 6227 + }, + { + "epoch": 0.9766311166875784, + "eval_scitail-pairs-pos_loss": 0.6142529249191284, + "eval_scitail-pairs-pos_runtime": 2.3525, + "eval_scitail-pairs-pos_samples_per_second": 85.018, + "eval_scitail-pairs-pos_steps_per_second": 2.125, + "step": 6227 + }, + { + "epoch": 0.9766311166875784, + "eval_xsum-pairs_loss": 0.5579215288162231, + "eval_xsum-pairs_runtime": 1.0381, + "eval_xsum-pairs_samples_per_second": 192.655, + "eval_xsum-pairs_steps_per_second": 4.816, + "step": 6227 + }, + { + "epoch": 0.9766311166875784, + "eval_compression-pairs_loss": 0.2538767158985138, + "eval_compression-pairs_runtime": 0.2367, + "eval_compression-pairs_samples_per_second": 844.99, + "eval_compression-pairs_steps_per_second": 21.125, + "step": 6227 + }, + { + "epoch": 0.9766311166875784, + "eval_sciq_pairs_loss": 0.43967145681381226, + "eval_sciq_pairs_runtime": 9.0489, + "eval_sciq_pairs_samples_per_second": 22.102, + "eval_sciq_pairs_steps_per_second": 0.553, + "step": 6227 + }, + { + "epoch": 0.9766311166875784, + "eval_qasc_pairs_loss": 0.7632485032081604, + "eval_qasc_pairs_runtime": 1.2216, + "eval_qasc_pairs_samples_per_second": 163.713, + "eval_qasc_pairs_steps_per_second": 4.093, + "step": 6227 + }, + { + "epoch": 0.9766311166875784, + "eval_openbookqa_pairs_loss": 2.370297908782959, + "eval_openbookqa_pairs_runtime": 1.051, + "eval_openbookqa_pairs_samples_per_second": 190.297, + "eval_openbookqa_pairs_steps_per_second": 4.757, + "step": 6227 + }, + { + "epoch": 0.9766311166875784, + "eval_msmarco_pairs_loss": 1.4191588163375854, + "eval_msmarco_pairs_runtime": 2.5294, + "eval_msmarco_pairs_samples_per_second": 79.071, + "eval_msmarco_pairs_steps_per_second": 1.977, + "step": 6227 + }, + { + "epoch": 0.9766311166875784, + "eval_nq_pairs_loss": 1.3847519159317017, + "eval_nq_pairs_runtime": 5.6464, + "eval_nq_pairs_samples_per_second": 35.421, + "eval_nq_pairs_steps_per_second": 0.886, + "step": 6227 + }, + { + "epoch": 0.9766311166875784, + "eval_trivia_pairs_loss": 1.6301060914993286, + "eval_trivia_pairs_runtime": 9.06, + "eval_trivia_pairs_samples_per_second": 22.075, + "eval_trivia_pairs_steps_per_second": 0.552, + "step": 6227 + }, + { + "epoch": 0.9766311166875784, + "eval_quora_pairs_loss": 0.25898078083992004, + "eval_quora_pairs_runtime": 0.6161, + "eval_quora_pairs_samples_per_second": 324.63, + "eval_quora_pairs_steps_per_second": 8.116, + "step": 6227 + }, + { + "epoch": 0.9766311166875784, + "eval_gooaq_pairs_loss": 1.1585972309112549, + "eval_gooaq_pairs_runtime": 1.547, + "eval_gooaq_pairs_samples_per_second": 129.286, + "eval_gooaq_pairs_steps_per_second": 3.232, + "step": 6227 + }, + { + "epoch": 0.9766311166875784, + "eval_mrpc_pairs_loss": 0.12749388813972473, + "eval_mrpc_pairs_runtime": 0.243, + "eval_mrpc_pairs_samples_per_second": 822.939, + "eval_mrpc_pairs_steps_per_second": 20.573, + "step": 6227 + }, + { + "epoch": 0.9786700125470514, + "grad_norm": 20.923500061035156, + "learning_rate": 3.157899806636098e-05, + "loss": 1.9813, + "step": 6240 + }, + { + "epoch": 0.986198243412798, + "grad_norm": 24.244462966918945, + "learning_rate": 3.13573957346308e-05, + "loss": 2.2932, + "step": 6288 + }, + { + "epoch": 0.9937264742785445, + "grad_norm": 16.324562072753906, + "learning_rate": 3.112966924777352e-05, + "loss": 1.6308, + "step": 6336 + }, + { + "epoch": 1.001254705144291, + "grad_norm": 14.824076652526855, + "learning_rate": 3.0895919247522884e-05, + "loss": 1.497, + "step": 6384 + }, + { + "epoch": 1.0087829360100375, + "grad_norm": 18.46307373046875, + "learning_rate": 3.065624903765184e-05, + "loss": 1.758, + "step": 6432 + }, + { + "epoch": 1.0163111668757843, + "grad_norm": 16.654727935791016, + "learning_rate": 3.0410764538318303e-05, + "loss": 1.6188, + "step": 6480 + }, + { + "epoch": 1.0238393977415308, + "grad_norm": 2.175520896911621, + "learning_rate": 3.0159574239254692e-05, + "loss": 2.1126, + "step": 6528 + }, + { + "epoch": 1.0313676286072773, + "grad_norm": 20.126880645751953, + "learning_rate": 2.990278915182182e-05, + "loss": 1.6129, + "step": 6576 + }, + { + "epoch": 1.0388958594730238, + "grad_norm": 24.355375289916992, + "learning_rate": 2.964052275994841e-05, + "loss": 1.4584, + "step": 6624 + }, + { + "epoch": 1.0464240903387705, + "grad_norm": 11.541900634765625, + "learning_rate": 2.9372890969977852e-05, + "loss": 1.5975, + "step": 6672 + }, + { + "epoch": 1.0517565872020076, + "eval_nli-pairs_loss": 1.3095624446868896, + "eval_nli-pairs_runtime": 4.0202, + "eval_nli-pairs_samples_per_second": 49.749, + "eval_nli-pairs_steps_per_second": 1.244, + "eval_sts-test_pearson_cosine": 0.7778731779702309, + "eval_sts-test_pearson_dot": 0.5593371858287987, + "eval_sts-test_pearson_euclidean": 0.7587435793478869, + "eval_sts-test_pearson_manhattan": 0.7610239287161901, + "eval_sts-test_pearson_max": 0.7778731779702309, + "eval_sts-test_spearman_cosine": 0.7782906514134392, + "eval_sts-test_spearman_dot": 0.5371251012528334, + "eval_sts-test_spearman_euclidean": 0.7452348039865185, + "eval_sts-test_spearman_manhattan": 0.7492626260972672, + "eval_sts-test_spearman_max": 0.7782906514134392, + "step": 6706 + }, + { + "epoch": 1.0517565872020076, + "eval_vitaminc-pairs_loss": 5.074347496032715, + "eval_vitaminc-pairs_runtime": 1.4418, + "eval_vitaminc-pairs_samples_per_second": 115.134, + "eval_vitaminc-pairs_steps_per_second": 2.774, + "step": 6706 + }, + { + "epoch": 1.0517565872020076, + "eval_sts-label_loss": 4.242542266845703, + "eval_sts-label_runtime": 0.414, + "eval_sts-label_samples_per_second": 483.143, + "eval_sts-label_steps_per_second": 12.079, + "step": 6706 + }, + { + "epoch": 1.0517565872020076, + "eval_qnli-contrastive_loss": 0.26250946521759033, + "eval_qnli-contrastive_runtime": 0.2857, + "eval_qnli-contrastive_samples_per_second": 700.035, + "eval_qnli-contrastive_steps_per_second": 17.501, + "step": 6706 + }, + { + "epoch": 1.0517565872020076, + "eval_scitail-pairs-qa_loss": 0.15288515388965607, + "eval_scitail-pairs-qa_runtime": 1.1069, + "eval_scitail-pairs-qa_samples_per_second": 180.69, + "eval_scitail-pairs-qa_steps_per_second": 4.517, + "step": 6706 + }, + { + "epoch": 1.0517565872020076, + "eval_scitail-pairs-pos_loss": 0.5382486581802368, + "eval_scitail-pairs-pos_runtime": 2.3648, + "eval_scitail-pairs-pos_samples_per_second": 84.574, + "eval_scitail-pairs-pos_steps_per_second": 2.114, + "step": 6706 + }, + { + "epoch": 1.0517565872020076, + "eval_xsum-pairs_loss": 0.48308631777763367, + "eval_xsum-pairs_runtime": 1.0411, + "eval_xsum-pairs_samples_per_second": 192.11, + "eval_xsum-pairs_steps_per_second": 4.803, + "step": 6706 + }, + { + "epoch": 1.0517565872020076, + "eval_compression-pairs_loss": 0.23988038301467896, + "eval_compression-pairs_runtime": 0.2389, + "eval_compression-pairs_samples_per_second": 837.159, + "eval_compression-pairs_steps_per_second": 20.929, + "step": 6706 + }, + { + "epoch": 1.0517565872020076, + "eval_sciq_pairs_loss": 0.3883107304573059, + "eval_sciq_pairs_runtime": 9.1094, + "eval_sciq_pairs_samples_per_second": 21.955, + "eval_sciq_pairs_steps_per_second": 0.549, + "step": 6706 + }, + { + "epoch": 1.0517565872020076, + "eval_qasc_pairs_loss": 0.6684954762458801, + "eval_qasc_pairs_runtime": 1.2249, + "eval_qasc_pairs_samples_per_second": 163.273, + "eval_qasc_pairs_steps_per_second": 4.082, + "step": 6706 + }, + { + "epoch": 1.0517565872020076, + "eval_openbookqa_pairs_loss": 2.076920986175537, + "eval_openbookqa_pairs_runtime": 1.0566, + "eval_openbookqa_pairs_samples_per_second": 189.291, + "eval_openbookqa_pairs_steps_per_second": 4.732, + "step": 6706 + }, + { + "epoch": 1.0517565872020076, + "eval_msmarco_pairs_loss": 1.3436123132705688, + "eval_msmarco_pairs_runtime": 2.5384, + "eval_msmarco_pairs_samples_per_second": 78.791, + "eval_msmarco_pairs_steps_per_second": 1.97, + "step": 6706 + }, + { + "epoch": 1.0517565872020076, + "eval_nq_pairs_loss": 1.2744060754776, + "eval_nq_pairs_runtime": 5.6694, + "eval_nq_pairs_samples_per_second": 35.277, + "eval_nq_pairs_steps_per_second": 0.882, + "step": 6706 + }, + { + "epoch": 1.0517565872020076, + "eval_trivia_pairs_loss": 1.356447458267212, + "eval_trivia_pairs_runtime": 9.0814, + "eval_trivia_pairs_samples_per_second": 22.023, + "eval_trivia_pairs_steps_per_second": 0.551, + "step": 6706 + }, + { + "epoch": 1.0517565872020076, + "eval_quora_pairs_loss": 0.20534881949424744, + "eval_quora_pairs_runtime": 0.6243, + "eval_quora_pairs_samples_per_second": 320.363, + "eval_quora_pairs_steps_per_second": 8.009, + "step": 6706 + }, + { + "epoch": 1.0517565872020076, + "eval_gooaq_pairs_loss": 1.131415843963623, + "eval_gooaq_pairs_runtime": 1.6004, + "eval_gooaq_pairs_samples_per_second": 124.966, + "eval_gooaq_pairs_steps_per_second": 3.124, + "step": 6706 + }, + { + "epoch": 1.0517565872020076, + "eval_mrpc_pairs_loss": 0.12877897918224335, + "eval_mrpc_pairs_runtime": 0.2425, + "eval_mrpc_pairs_samples_per_second": 824.742, + "eval_mrpc_pairs_steps_per_second": 20.619, + "step": 6706 + }, + { + "epoch": 1.053952321204517, + "grad_norm": 12.653462409973145, + "learning_rate": 2.9100012059444395e-05, + "loss": 1.6933, + "step": 6720 + }, + { + "epoch": 1.0614805520702635, + "grad_norm": 13.93374252319336, + "learning_rate": 2.8822006624801445e-05, + "loss": 1.2931, + "step": 6768 + }, + { + "epoch": 1.06900878293601, + "grad_norm": 17.680423736572266, + "learning_rate": 2.8538997528125016e-05, + "loss": 1.8077, + "step": 6816 + }, + { + "epoch": 1.0765370138017567, + "grad_norm": 14.7294921875, + "learning_rate": 2.8251109842815857e-05, + "loss": 1.4748, + "step": 6864 + }, + { + "epoch": 1.0840652446675032, + "grad_norm": 26.121625900268555, + "learning_rate": 2.795847079832445e-05, + "loss": 1.225, + "step": 6912 + }, + { + "epoch": 1.0915934755332497, + "grad_norm": 5.665552616119385, + "learning_rate": 2.7661209723922966e-05, + "loss": 1.7627, + "step": 6960 + }, + { + "epoch": 1.0991217063989962, + "grad_norm": 1.2186944484710693, + "learning_rate": 2.7359457991549365e-05, + "loss": 1.5613, + "step": 7008 + }, + { + "epoch": 1.1066499372647427, + "grad_norm": 4.569064617156982, + "learning_rate": 2.7053348957748744e-05, + "loss": 1.6548, + "step": 7056 + }, + { + "epoch": 1.1141781681304894, + "grad_norm": 11.330352783203125, + "learning_rate": 2.6743017904737505e-05, + "loss": 1.4033, + "step": 7104 + }, + { + "epoch": 1.121706398996236, + "grad_norm": 27.65351676940918, + "learning_rate": 2.6428601980616635e-05, + "loss": 1.3999, + "step": 7152 + }, + { + "epoch": 1.1268820577164367, + "eval_nli-pairs_loss": 1.212507724761963, + "eval_nli-pairs_runtime": 4.0093, + "eval_nli-pairs_samples_per_second": 49.884, + "eval_nli-pairs_steps_per_second": 1.247, + "eval_sts-test_pearson_cosine": 0.771726417661154, + "eval_sts-test_pearson_dot": 0.5582118135503709, + "eval_sts-test_pearson_euclidean": 0.7582436737188462, + "eval_sts-test_pearson_manhattan": 0.7602002352027541, + "eval_sts-test_pearson_max": 0.771726417661154, + "eval_sts-test_spearman_cosine": 0.7760137659698252, + "eval_sts-test_spearman_dot": 0.5314151368436631, + "eval_sts-test_spearman_euclidean": 0.7447358001087161, + "eval_sts-test_spearman_manhattan": 0.7487681272112986, + "eval_sts-test_spearman_max": 0.7760137659698252, + "step": 7185 + }, + { + "epoch": 1.1268820577164367, + "eval_vitaminc-pairs_loss": 4.9936699867248535, + "eval_vitaminc-pairs_runtime": 1.4575, + "eval_vitaminc-pairs_samples_per_second": 113.89, + "eval_vitaminc-pairs_steps_per_second": 2.744, + "step": 7185 + }, + { + "epoch": 1.1268820577164367, + "eval_sts-label_loss": 4.1703057289123535, + "eval_sts-label_runtime": 0.4168, + "eval_sts-label_samples_per_second": 479.874, + "eval_sts-label_steps_per_second": 11.997, + "step": 7185 + }, + { + "epoch": 1.1268820577164367, + "eval_qnli-contrastive_loss": 0.31846657395362854, + "eval_qnli-contrastive_runtime": 0.2838, + "eval_qnli-contrastive_samples_per_second": 704.663, + "eval_qnli-contrastive_steps_per_second": 17.617, + "step": 7185 + }, + { + "epoch": 1.1268820577164367, + "eval_scitail-pairs-qa_loss": 0.1417759656906128, + "eval_scitail-pairs-qa_runtime": 1.0581, + "eval_scitail-pairs-qa_samples_per_second": 189.022, + "eval_scitail-pairs-qa_steps_per_second": 4.726, + "step": 7185 + }, + { + "epoch": 1.1268820577164367, + "eval_scitail-pairs-pos_loss": 0.5202088356018066, + "eval_scitail-pairs-pos_runtime": 2.4081, + "eval_scitail-pairs-pos_samples_per_second": 83.054, + "eval_scitail-pairs-pos_steps_per_second": 2.076, + "step": 7185 + }, + { + "epoch": 1.1268820577164367, + "eval_xsum-pairs_loss": 0.44685080647468567, + "eval_xsum-pairs_runtime": 1.0416, + "eval_xsum-pairs_samples_per_second": 192.009, + "eval_xsum-pairs_steps_per_second": 4.8, + "step": 7185 + }, + { + "epoch": 1.1268820577164367, + "eval_compression-pairs_loss": 0.22411338984966278, + "eval_compression-pairs_runtime": 0.2438, + "eval_compression-pairs_samples_per_second": 820.201, + "eval_compression-pairs_steps_per_second": 20.505, + "step": 7185 + }, + { + "epoch": 1.1268820577164367, + "eval_sciq_pairs_loss": 0.37388285994529724, + "eval_sciq_pairs_runtime": 9.0875, + "eval_sciq_pairs_samples_per_second": 22.008, + "eval_sciq_pairs_steps_per_second": 0.55, + "step": 7185 + }, + { + "epoch": 1.1268820577164367, + "eval_qasc_pairs_loss": 0.6410768032073975, + "eval_qasc_pairs_runtime": 1.2169, + "eval_qasc_pairs_samples_per_second": 164.348, + "eval_qasc_pairs_steps_per_second": 4.109, + "step": 7185 + }, + { + "epoch": 1.1268820577164367, + "eval_openbookqa_pairs_loss": 2.043297052383423, + "eval_openbookqa_pairs_runtime": 1.0552, + "eval_openbookqa_pairs_samples_per_second": 189.539, + "eval_openbookqa_pairs_steps_per_second": 4.738, + "step": 7185 + }, + { + "epoch": 1.1268820577164367, + "eval_msmarco_pairs_loss": 1.2795078754425049, + "eval_msmarco_pairs_runtime": 2.5343, + "eval_msmarco_pairs_samples_per_second": 78.919, + "eval_msmarco_pairs_steps_per_second": 1.973, + "step": 7185 + }, + { + "epoch": 1.1268820577164367, + "eval_nq_pairs_loss": 1.1771754026412964, + "eval_nq_pairs_runtime": 5.6589, + "eval_nq_pairs_samples_per_second": 35.343, + "eval_nq_pairs_steps_per_second": 0.884, + "step": 7185 + }, + { + "epoch": 1.1268820577164367, + "eval_trivia_pairs_loss": 1.305577039718628, + "eval_trivia_pairs_runtime": 9.0872, + "eval_trivia_pairs_samples_per_second": 22.009, + "eval_trivia_pairs_steps_per_second": 0.55, + "step": 7185 + }, + { + "epoch": 1.1268820577164367, + "eval_quora_pairs_loss": 0.2645386755466461, + "eval_quora_pairs_runtime": 0.6429, + "eval_quora_pairs_samples_per_second": 311.086, + "eval_quora_pairs_steps_per_second": 7.777, + "step": 7185 + }, + { + "epoch": 1.1268820577164367, + "eval_gooaq_pairs_loss": 1.0562756061553955, + "eval_gooaq_pairs_runtime": 1.5772, + "eval_gooaq_pairs_samples_per_second": 126.809, + "eval_gooaq_pairs_steps_per_second": 3.17, + "step": 7185 + }, + { + "epoch": 1.1268820577164367, + "eval_mrpc_pairs_loss": 0.1197453960776329, + "eval_mrpc_pairs_runtime": 0.2393, + "eval_mrpc_pairs_samples_per_second": 835.714, + "eval_mrpc_pairs_steps_per_second": 20.893, + "step": 7185 + }, + { + "epoch": 1.1292346298619824, + "grad_norm": 26.328977584838867, + "learning_rate": 2.611024013876024e-05, + "loss": 1.707, + "step": 7200 + }, + { + "epoch": 1.136762860727729, + "grad_norm": 14.730170249938965, + "learning_rate": 2.578807307640633e-05, + "loss": 1.2812, + "step": 7248 + }, + { + "epoch": 1.1442910915934754, + "grad_norm": 16.989944458007812, + "learning_rate": 2.546224317247695e-05, + "loss": 2.0209, + "step": 7296 + }, + { + "epoch": 1.1518193224592221, + "grad_norm": 15.942720413208008, + "learning_rate": 2.5132894424654982e-05, + "loss": 1.465, + "step": 7344 + }, + { + "epoch": 1.1593475533249686, + "grad_norm": 13.42844009399414, + "learning_rate": 2.4800172385745665e-05, + "loss": 1.6181, + "step": 7392 + }, + { + "epoch": 1.1668757841907151, + "grad_norm": 2.6255691051483154, + "learning_rate": 2.446422409935082e-05, + "loss": 1.3386, + "step": 7440 + }, + { + "epoch": 1.1744040150564616, + "grad_norm": 11.230644226074219, + "learning_rate": 2.412519803488417e-05, + "loss": 1.4015, + "step": 7488 + }, + { + "epoch": 1.1819322459222084, + "grad_norm": 17.679582595825195, + "learning_rate": 2.3783244021956605e-05, + "loss": 1.5057, + "step": 7536 + }, + { + "epoch": 1.1894604767879549, + "grad_norm": 12.440255165100098, + "learning_rate": 2.3438513184160302e-05, + "loss": 1.2441, + "step": 7584 + }, + { + "epoch": 1.1969887076537014, + "grad_norm": 17.67916488647461, + "learning_rate": 2.309115787228094e-05, + "loss": 1.1718, + "step": 7632 + }, + { + "epoch": 1.2020075282308658, + "eval_nli-pairs_loss": 1.1914833784103394, + "eval_nli-pairs_runtime": 4.1213, + "eval_nli-pairs_samples_per_second": 48.528, + "eval_nli-pairs_steps_per_second": 1.213, + "eval_sts-test_pearson_cosine": 0.7800804097388158, + "eval_sts-test_pearson_dot": 0.5541152451346555, + "eval_sts-test_pearson_euclidean": 0.7555650256629914, + "eval_sts-test_pearson_manhattan": 0.7570661501424889, + "eval_sts-test_pearson_max": 0.7800804097388158, + "eval_sts-test_spearman_cosine": 0.7837125392806782, + "eval_sts-test_spearman_dot": 0.5299991823870257, + "eval_sts-test_spearman_euclidean": 0.7420646288254193, + "eval_sts-test_spearman_manhattan": 0.7454823370503758, + "eval_sts-test_spearman_max": 0.7837125392806782, + "step": 7664 + }, + { + "epoch": 1.2020075282308658, + "eval_vitaminc-pairs_loss": 4.815649509429932, + "eval_vitaminc-pairs_runtime": 1.4965, + "eval_vitaminc-pairs_samples_per_second": 110.928, + "eval_vitaminc-pairs_steps_per_second": 2.673, + "step": 7664 + }, + { + "epoch": 1.2020075282308658, + "eval_sts-label_loss": 4.200761795043945, + "eval_sts-label_runtime": 0.4104, + "eval_sts-label_samples_per_second": 487.308, + "eval_sts-label_steps_per_second": 12.183, + "step": 7664 + }, + { + "epoch": 1.2020075282308658, + "eval_qnli-contrastive_loss": 0.21287916600704193, + "eval_qnli-contrastive_runtime": 0.2849, + "eval_qnli-contrastive_samples_per_second": 702.086, + "eval_qnli-contrastive_steps_per_second": 17.552, + "step": 7664 + }, + { + "epoch": 1.2020075282308658, + "eval_scitail-pairs-qa_loss": 0.12968133389949799, + "eval_scitail-pairs-qa_runtime": 1.0998, + "eval_scitail-pairs-qa_samples_per_second": 181.845, + "eval_scitail-pairs-qa_steps_per_second": 4.546, + "step": 7664 + }, + { + "epoch": 1.2020075282308658, + "eval_scitail-pairs-pos_loss": 0.4889526665210724, + "eval_scitail-pairs-pos_runtime": 2.3956, + "eval_scitail-pairs-pos_samples_per_second": 83.486, + "eval_scitail-pairs-pos_steps_per_second": 2.087, + "step": 7664 + }, + { + "epoch": 1.2020075282308658, + "eval_xsum-pairs_loss": 0.42897719144821167, + "eval_xsum-pairs_runtime": 1.0492, + "eval_xsum-pairs_samples_per_second": 190.623, + "eval_xsum-pairs_steps_per_second": 4.766, + "step": 7664 + }, + { + "epoch": 1.2020075282308658, + "eval_compression-pairs_loss": 0.20111607015132904, + "eval_compression-pairs_runtime": 0.2549, + "eval_compression-pairs_samples_per_second": 784.74, + "eval_compression-pairs_steps_per_second": 19.618, + "step": 7664 + }, + { + "epoch": 1.2020075282308658, + "eval_sciq_pairs_loss": 0.36706623435020447, + "eval_sciq_pairs_runtime": 9.1959, + "eval_sciq_pairs_samples_per_second": 21.749, + "eval_sciq_pairs_steps_per_second": 0.544, + "step": 7664 + }, + { + "epoch": 1.2020075282308658, + "eval_qasc_pairs_loss": 0.5133731961250305, + "eval_qasc_pairs_runtime": 1.2209, + "eval_qasc_pairs_samples_per_second": 163.813, + "eval_qasc_pairs_steps_per_second": 4.095, + "step": 7664 + }, + { + "epoch": 1.2020075282308658, + "eval_openbookqa_pairs_loss": 1.9764889478683472, + "eval_openbookqa_pairs_runtime": 1.1177, + "eval_openbookqa_pairs_samples_per_second": 178.939, + "eval_openbookqa_pairs_steps_per_second": 4.473, + "step": 7664 + }, + { + "epoch": 1.2020075282308658, + "eval_msmarco_pairs_loss": 1.1972062587738037, + "eval_msmarco_pairs_runtime": 2.5812, + "eval_msmarco_pairs_samples_per_second": 77.485, + "eval_msmarco_pairs_steps_per_second": 1.937, + "step": 7664 + }, + { + "epoch": 1.2020075282308658, + "eval_nq_pairs_loss": 1.1358052492141724, + "eval_nq_pairs_runtime": 5.7952, + "eval_nq_pairs_samples_per_second": 34.511, + "eval_nq_pairs_steps_per_second": 0.863, + "step": 7664 + }, + { + "epoch": 1.2020075282308658, + "eval_trivia_pairs_loss": 1.3059653043746948, + "eval_trivia_pairs_runtime": 9.1651, + "eval_trivia_pairs_samples_per_second": 21.822, + "eval_trivia_pairs_steps_per_second": 0.546, + "step": 7664 + }, + { + "epoch": 1.2020075282308658, + "eval_quora_pairs_loss": 0.23538345098495483, + "eval_quora_pairs_runtime": 0.6553, + "eval_quora_pairs_samples_per_second": 305.181, + "eval_quora_pairs_steps_per_second": 7.63, + "step": 7664 + }, + { + "epoch": 1.2020075282308658, + "eval_gooaq_pairs_loss": 1.0195775032043457, + "eval_gooaq_pairs_runtime": 1.5615, + "eval_gooaq_pairs_samples_per_second": 128.081, + "eval_gooaq_pairs_steps_per_second": 3.202, + "step": 7664 + }, + { + "epoch": 1.2020075282308658, + "eval_mrpc_pairs_loss": 0.10825151205062866, + "eval_mrpc_pairs_runtime": 0.2544, + "eval_mrpc_pairs_samples_per_second": 786.113, + "eval_mrpc_pairs_steps_per_second": 19.653, + "step": 7664 + }, + { + "epoch": 1.2045169385194479, + "grad_norm": 11.216833114624023, + "learning_rate": 2.2741331596967632e-05, + "loss": 1.7562, + "step": 7680 + }, + { + "epoch": 1.2120451693851946, + "grad_norm": 0.7245489358901978, + "learning_rate": 2.238918896089022e-05, + "loss": 1.2954, + "step": 7728 + }, + { + "epoch": 1.219573400250941, + "grad_norm": 18.958086013793945, + "learning_rate": 2.2034885590413966e-05, + "loss": 1.7821, + "step": 7776 + }, + { + "epoch": 1.2271016311166876, + "grad_norm": 1.1195874214172363, + "learning_rate": 2.1678578066821842e-05, + "loss": 1.4494, + "step": 7824 + }, + { + "epoch": 1.234629861982434, + "grad_norm": 10.63464641571045, + "learning_rate": 2.1320423857114807e-05, + "loss": 1.1674, + "step": 7872 + }, + { + "epoch": 1.2421580928481806, + "grad_norm": 20.8350830078125, + "learning_rate": 2.0960581244420636e-05, + "loss": 1.1511, + "step": 7920 + }, + { + "epoch": 1.2496863237139273, + "grad_norm": 20.231365203857422, + "learning_rate": 2.0599209258042067e-05, + "loss": 1.1629, + "step": 7968 + }, + { + "epoch": 1.2572145545796738, + "grad_norm": 2.3765740394592285, + "learning_rate": 2.023646760317521e-05, + "loss": 1.167, + "step": 8016 + }, + { + "epoch": 1.2647427854454203, + "grad_norm": 0.45109543204307556, + "learning_rate": 1.9872516590329177e-05, + "loss": 1.5169, + "step": 8064 + }, + { + "epoch": 1.272271016311167, + "grad_norm": 19.704111099243164, + "learning_rate": 1.9507517064478322e-05, + "loss": 1.1595, + "step": 8112 + }, + { + "epoch": 1.2771329987452948, + "eval_nli-pairs_loss": 1.117584466934204, + "eval_nli-pairs_runtime": 4.0331, + "eval_nli-pairs_samples_per_second": 49.59, + "eval_nli-pairs_steps_per_second": 1.24, + "eval_sts-test_pearson_cosine": 0.7735358515290777, + "eval_sts-test_pearson_dot": 0.560523750110175, + "eval_sts-test_pearson_euclidean": 0.7469529952792658, + "eval_sts-test_pearson_manhattan": 0.7479614812850265, + "eval_sts-test_pearson_max": 0.7735358515290777, + "eval_sts-test_spearman_cosine": 0.7789103785997893, + "eval_sts-test_spearman_dot": 0.534930979573433, + "eval_sts-test_spearman_euclidean": 0.7320475561187462, + "eval_sts-test_spearman_manhattan": 0.7359660045883268, + "eval_sts-test_spearman_max": 0.7789103785997893, + "step": 8143 + }, + { + "epoch": 1.2771329987452948, + "eval_vitaminc-pairs_loss": 4.841164588928223, + "eval_vitaminc-pairs_runtime": 1.4299, + "eval_vitaminc-pairs_samples_per_second": 116.092, + "eval_vitaminc-pairs_steps_per_second": 2.797, + "step": 8143 + }, + { + "epoch": 1.2771329987452948, + "eval_sts-label_loss": 4.0971784591674805, + "eval_sts-label_runtime": 0.4026, + "eval_sts-label_samples_per_second": 496.749, + "eval_sts-label_steps_per_second": 12.419, + "step": 8143 + }, + { + "epoch": 1.2771329987452948, + "eval_qnli-contrastive_loss": 0.2050691843032837, + "eval_qnli-contrastive_runtime": 0.2817, + "eval_qnli-contrastive_samples_per_second": 710.023, + "eval_qnli-contrastive_steps_per_second": 17.751, + "step": 8143 + }, + { + "epoch": 1.2771329987452948, + "eval_scitail-pairs-qa_loss": 0.11168017983436584, + "eval_scitail-pairs-qa_runtime": 1.056, + "eval_scitail-pairs-qa_samples_per_second": 189.398, + "eval_scitail-pairs-qa_steps_per_second": 4.735, + "step": 8143 + }, + { + "epoch": 1.2771329987452948, + "eval_scitail-pairs-pos_loss": 0.495766818523407, + "eval_scitail-pairs-pos_runtime": 2.4272, + "eval_scitail-pairs-pos_samples_per_second": 82.4, + "eval_scitail-pairs-pos_steps_per_second": 2.06, + "step": 8143 + }, + { + "epoch": 1.2771329987452948, + "eval_xsum-pairs_loss": 0.43334144353866577, + "eval_xsum-pairs_runtime": 1.0522, + "eval_xsum-pairs_samples_per_second": 190.086, + "eval_xsum-pairs_steps_per_second": 4.752, + "step": 8143 + }, + { + "epoch": 1.2771329987452948, + "eval_compression-pairs_loss": 0.1837506741285324, + "eval_compression-pairs_runtime": 0.2435, + "eval_compression-pairs_samples_per_second": 821.487, + "eval_compression-pairs_steps_per_second": 20.537, + "step": 8143 + }, + { + "epoch": 1.2771329987452948, + "eval_sciq_pairs_loss": 0.35108524560928345, + "eval_sciq_pairs_runtime": 9.102, + "eval_sciq_pairs_samples_per_second": 21.973, + "eval_sciq_pairs_steps_per_second": 0.549, + "step": 8143 + }, + { + "epoch": 1.2771329987452948, + "eval_qasc_pairs_loss": 0.486527681350708, + "eval_qasc_pairs_runtime": 1.2312, + "eval_qasc_pairs_samples_per_second": 162.45, + "eval_qasc_pairs_steps_per_second": 4.061, + "step": 8143 + }, + { + "epoch": 1.2771329987452948, + "eval_openbookqa_pairs_loss": 1.9944250583648682, + "eval_openbookqa_pairs_runtime": 1.0997, + "eval_openbookqa_pairs_samples_per_second": 181.872, + "eval_openbookqa_pairs_steps_per_second": 4.547, + "step": 8143 + }, + { + "epoch": 1.2771329987452948, + "eval_msmarco_pairs_loss": 1.1277974843978882, + "eval_msmarco_pairs_runtime": 2.5357, + "eval_msmarco_pairs_samples_per_second": 78.874, + "eval_msmarco_pairs_steps_per_second": 1.972, + "step": 8143 + }, + { + "epoch": 1.2771329987452948, + "eval_nq_pairs_loss": 1.092708945274353, + "eval_nq_pairs_runtime": 5.6524, + "eval_nq_pairs_samples_per_second": 35.383, + "eval_nq_pairs_steps_per_second": 0.885, + "step": 8143 + }, + { + "epoch": 1.2771329987452948, + "eval_trivia_pairs_loss": 1.3160641193389893, + "eval_trivia_pairs_runtime": 9.0422, + "eval_trivia_pairs_samples_per_second": 22.118, + "eval_trivia_pairs_steps_per_second": 0.553, + "step": 8143 + }, + { + "epoch": 1.2771329987452948, + "eval_quora_pairs_loss": 0.1377909630537033, + "eval_quora_pairs_runtime": 0.6252, + "eval_quora_pairs_samples_per_second": 319.913, + "eval_quora_pairs_steps_per_second": 7.998, + "step": 8143 + }, + { + "epoch": 1.2771329987452948, + "eval_gooaq_pairs_loss": 0.9788944721221924, + "eval_gooaq_pairs_runtime": 1.5536, + "eval_gooaq_pairs_samples_per_second": 128.731, + "eval_gooaq_pairs_steps_per_second": 3.218, + "step": 8143 + }, + { + "epoch": 1.2771329987452948, + "eval_mrpc_pairs_loss": 0.09496909379959106, + "eval_mrpc_pairs_runtime": 0.2386, + "eval_mrpc_pairs_samples_per_second": 838.384, + "eval_mrpc_pairs_steps_per_second": 20.96, + "step": 8143 + }, + { + "epoch": 1.2797992471769133, + "grad_norm": 10.626935958862305, + "learning_rate": 1.914163033397814e-05, + "loss": 1.4598, + "step": 8160 + }, + { + "epoch": 1.28732747804266, + "grad_norm": 70.21989440917969, + "learning_rate": 1.8775018099276515e-05, + "loss": 1.2038, + "step": 8208 + }, + { + "epoch": 1.2948557089084065, + "grad_norm": 13.136975288391113, + "learning_rate": 1.8407842381451562e-05, + "loss": 1.2381, + "step": 8256 + }, + { + "epoch": 1.302383939774153, + "grad_norm": 14.311907768249512, + "learning_rate": 1.804026545060783e-05, + "loss": 0.9591, + "step": 8304 + }, + { + "epoch": 1.3099121706398997, + "grad_norm": 11.372726440429688, + "learning_rate": 1.767244975416249e-05, + "loss": 1.3451, + "step": 8352 + }, + { + "epoch": 1.3174404015056462, + "grad_norm": 12.1253080368042, + "learning_rate": 1.7304557845053075e-05, + "loss": 1.2669, + "step": 8400 + }, + { + "epoch": 1.3249686323713927, + "grad_norm": 10.53577995300293, + "learning_rate": 1.693675230989863e-05, + "loss": 1.4754, + "step": 8448 + }, + { + "epoch": 1.3324968632371392, + "grad_norm": 16.669851303100586, + "learning_rate": 1.6576849494355775e-05, + "loss": 1.2094, + "step": 8496 + }, + { + "epoch": 1.3400250941028857, + "grad_norm": 0.5680477023124695, + "learning_rate": 1.6209694016709658e-05, + "loss": 1.4406, + "step": 8544 + }, + { + "epoch": 1.3475533249686324, + "grad_norm": 13.915863990783691, + "learning_rate": 1.5843108778533816e-05, + "loss": 1.3063, + "step": 8592 + }, + { + "epoch": 1.352258469259724, + "eval_nli-pairs_loss": 1.1088637113571167, + "eval_nli-pairs_runtime": 4.0091, + "eval_nli-pairs_samples_per_second": 49.887, + "eval_nli-pairs_steps_per_second": 1.247, + "eval_sts-test_pearson_cosine": 0.7694392469514715, + "eval_sts-test_pearson_dot": 0.5341434503970798, + "eval_sts-test_pearson_euclidean": 0.7443691249920426, + "eval_sts-test_pearson_manhattan": 0.7461066741203818, + "eval_sts-test_pearson_max": 0.7694392469514715, + "eval_sts-test_spearman_cosine": 0.7756586193959978, + "eval_sts-test_spearman_dot": 0.5051721749088769, + "eval_sts-test_spearman_euclidean": 0.73008171628055, + "eval_sts-test_spearman_manhattan": 0.7347429927090328, + "eval_sts-test_spearman_max": 0.7756586193959978, + "step": 8622 + }, + { + "epoch": 1.352258469259724, + "eval_vitaminc-pairs_loss": 4.769205093383789, + "eval_vitaminc-pairs_runtime": 1.4375, + "eval_vitaminc-pairs_samples_per_second": 115.477, + "eval_vitaminc-pairs_steps_per_second": 2.783, + "step": 8622 + }, + { + "epoch": 1.352258469259724, + "eval_sts-label_loss": 4.07755184173584, + "eval_sts-label_runtime": 0.4049, + "eval_sts-label_samples_per_second": 493.895, + "eval_sts-label_steps_per_second": 12.347, + "step": 8622 + }, + { + "epoch": 1.352258469259724, + "eval_qnli-contrastive_loss": 0.17923730611801147, + "eval_qnli-contrastive_runtime": 0.2806, + "eval_qnli-contrastive_samples_per_second": 712.745, + "eval_qnli-contrastive_steps_per_second": 17.819, + "step": 8622 + }, + { + "epoch": 1.352258469259724, + "eval_scitail-pairs-qa_loss": 0.10800071805715561, + "eval_scitail-pairs-qa_runtime": 1.1053, + "eval_scitail-pairs-qa_samples_per_second": 180.952, + "eval_scitail-pairs-qa_steps_per_second": 4.524, + "step": 8622 + }, + { + "epoch": 1.352258469259724, + "eval_scitail-pairs-pos_loss": 0.48035934567451477, + "eval_scitail-pairs-pos_runtime": 2.3848, + "eval_scitail-pairs-pos_samples_per_second": 83.864, + "eval_scitail-pairs-pos_steps_per_second": 2.097, + "step": 8622 + }, + { + "epoch": 1.352258469259724, + "eval_xsum-pairs_loss": 0.408091276884079, + "eval_xsum-pairs_runtime": 1.043, + "eval_xsum-pairs_samples_per_second": 191.746, + "eval_xsum-pairs_steps_per_second": 4.794, + "step": 8622 + }, + { + "epoch": 1.352258469259724, + "eval_compression-pairs_loss": 0.17393065989017487, + "eval_compression-pairs_runtime": 0.2388, + "eval_compression-pairs_samples_per_second": 837.593, + "eval_compression-pairs_steps_per_second": 20.94, + "step": 8622 + }, + { + "epoch": 1.352258469259724, + "eval_sciq_pairs_loss": 0.3497403562068939, + "eval_sciq_pairs_runtime": 9.1128, + "eval_sciq_pairs_samples_per_second": 21.947, + "eval_sciq_pairs_steps_per_second": 0.549, + "step": 8622 + }, + { + "epoch": 1.352258469259724, + "eval_qasc_pairs_loss": 0.4620342552661896, + "eval_qasc_pairs_runtime": 1.2208, + "eval_qasc_pairs_samples_per_second": 163.822, + "eval_qasc_pairs_steps_per_second": 4.096, + "step": 8622 + }, + { + "epoch": 1.352258469259724, + "eval_openbookqa_pairs_loss": 1.9002811908721924, + "eval_openbookqa_pairs_runtime": 1.0593, + "eval_openbookqa_pairs_samples_per_second": 188.81, + "eval_openbookqa_pairs_steps_per_second": 4.72, + "step": 8622 + }, + { + "epoch": 1.352258469259724, + "eval_msmarco_pairs_loss": 1.0928370952606201, + "eval_msmarco_pairs_runtime": 2.5227, + "eval_msmarco_pairs_samples_per_second": 79.281, + "eval_msmarco_pairs_steps_per_second": 1.982, + "step": 8622 + }, + { + "epoch": 1.352258469259724, + "eval_nq_pairs_loss": 1.0306727886199951, + "eval_nq_pairs_runtime": 5.6685, + "eval_nq_pairs_samples_per_second": 35.283, + "eval_nq_pairs_steps_per_second": 0.882, + "step": 8622 + }, + { + "epoch": 1.352258469259724, + "eval_trivia_pairs_loss": 1.3031796216964722, + "eval_trivia_pairs_runtime": 9.0827, + "eval_trivia_pairs_samples_per_second": 22.02, + "eval_trivia_pairs_steps_per_second": 0.55, + "step": 8622 + }, + { + "epoch": 1.352258469259724, + "eval_quora_pairs_loss": 0.16622459888458252, + "eval_quora_pairs_runtime": 0.6233, + "eval_quora_pairs_samples_per_second": 320.863, + "eval_quora_pairs_steps_per_second": 8.022, + "step": 8622 + }, + { + "epoch": 1.352258469259724, + "eval_gooaq_pairs_loss": 0.9280920624732971, + "eval_gooaq_pairs_runtime": 1.5586, + "eval_gooaq_pairs_samples_per_second": 128.317, + "eval_gooaq_pairs_steps_per_second": 3.208, + "step": 8622 + }, + { + "epoch": 1.352258469259724, + "eval_mrpc_pairs_loss": 0.09042897820472717, + "eval_mrpc_pairs_runtime": 0.2477, + "eval_mrpc_pairs_samples_per_second": 807.373, + "eval_mrpc_pairs_steps_per_second": 20.184, + "step": 8622 + }, + { + "epoch": 1.355081555834379, + "grad_norm": 10.317242622375488, + "learning_rate": 1.5477255788975485e-05, + "loss": 1.1662, + "step": 8640 + }, + { + "epoch": 1.3626097867001254, + "grad_norm": 14.411606788635254, + "learning_rate": 1.5112296733571015e-05, + "loss": 1.1592, + "step": 8688 + }, + { + "epoch": 1.370138017565872, + "grad_norm": 14.328046798706055, + "learning_rate": 1.474839290279036e-05, + "loss": 1.1742, + "step": 8736 + }, + { + "epoch": 1.3776662484316184, + "grad_norm": 15.689784049987793, + "learning_rate": 1.4385705120756155e-05, + "loss": 1.2304, + "step": 8784 + }, + { + "epoch": 1.3851944792973652, + "grad_norm": 1.1622222661972046, + "learning_rate": 1.4024393674168905e-05, + "loss": 1.0637, + "step": 8832 + }, + { + "epoch": 1.3927227101631117, + "grad_norm": 55.88763427734375, + "learning_rate": 1.3664618241469671e-05, + "loss": 1.8133, + "step": 8880 + }, + { + "epoch": 1.4002509410288582, + "grad_norm": 11.148764610290527, + "learning_rate": 1.3306537822271584e-05, + "loss": 1.682, + "step": 8928 + }, + { + "epoch": 1.4077791718946049, + "grad_norm": 34.30131149291992, + "learning_rate": 1.2950310667091337e-05, + "loss": 1.1823, + "step": 8976 + }, + { + "epoch": 1.4153074027603514, + "grad_norm": 1.3407506942749023, + "learning_rate": 1.2596094207411785e-05, + "loss": 1.2953, + "step": 9024 + }, + { + "epoch": 1.4228356336260979, + "grad_norm": 0.3961174190044403, + "learning_rate": 1.2244044986106427e-05, + "loss": 0.9769, + "step": 9072 + }, + { + "epoch": 1.427383939774153, + "eval_nli-pairs_loss": 1.065342664718628, + "eval_nli-pairs_runtime": 4.0707, + "eval_nli-pairs_samples_per_second": 49.132, + "eval_nli-pairs_steps_per_second": 1.228, + "eval_sts-test_pearson_cosine": 0.7731513422471518, + "eval_sts-test_pearson_dot": 0.5353916353979121, + "eval_sts-test_pearson_euclidean": 0.7469853367274143, + "eval_sts-test_pearson_manhattan": 0.7481039749789392, + "eval_sts-test_pearson_max": 0.7731513422471518, + "eval_sts-test_spearman_cosine": 0.7792255681202213, + "eval_sts-test_spearman_dot": 0.5083392134944579, + "eval_sts-test_spearman_euclidean": 0.7335576317363027, + "eval_sts-test_spearman_manhattan": 0.7370337922992453, + "eval_sts-test_spearman_max": 0.7792255681202213, + "step": 9101 + }, + { + "epoch": 1.427383939774153, + "eval_vitaminc-pairs_loss": 4.8438720703125, + "eval_vitaminc-pairs_runtime": 1.4283, + "eval_vitaminc-pairs_samples_per_second": 116.218, + "eval_vitaminc-pairs_steps_per_second": 2.8, + "step": 9101 + }, + { + "epoch": 1.427383939774153, + "eval_sts-label_loss": 4.092858791351318, + "eval_sts-label_runtime": 0.407, + "eval_sts-label_samples_per_second": 491.445, + "eval_sts-label_steps_per_second": 12.286, + "step": 9101 + }, + { + "epoch": 1.427383939774153, + "eval_qnli-contrastive_loss": 0.15653812885284424, + "eval_qnli-contrastive_runtime": 0.2813, + "eval_qnli-contrastive_samples_per_second": 711.049, + "eval_qnli-contrastive_steps_per_second": 17.776, + "step": 9101 + }, + { + "epoch": 1.427383939774153, + "eval_scitail-pairs-qa_loss": 0.10110773146152496, + "eval_scitail-pairs-qa_runtime": 1.0598, + "eval_scitail-pairs-qa_samples_per_second": 188.718, + "eval_scitail-pairs-qa_steps_per_second": 4.718, + "step": 9101 + }, + { + "epoch": 1.427383939774153, + "eval_scitail-pairs-pos_loss": 0.46982836723327637, + "eval_scitail-pairs-pos_runtime": 2.3471, + "eval_scitail-pairs-pos_samples_per_second": 85.213, + "eval_scitail-pairs-pos_steps_per_second": 2.13, + "step": 9101 + }, + { + "epoch": 1.427383939774153, + "eval_xsum-pairs_loss": 0.3937150537967682, + "eval_xsum-pairs_runtime": 1.0402, + "eval_xsum-pairs_samples_per_second": 192.278, + "eval_xsum-pairs_steps_per_second": 4.807, + "step": 9101 + }, + { + "epoch": 1.427383939774153, + "eval_compression-pairs_loss": 0.16437682509422302, + "eval_compression-pairs_runtime": 0.2405, + "eval_compression-pairs_samples_per_second": 831.709, + "eval_compression-pairs_steps_per_second": 20.793, + "step": 9101 + }, + { + "epoch": 1.427383939774153, + "eval_sciq_pairs_loss": 0.3407261073589325, + "eval_sciq_pairs_runtime": 9.0549, + "eval_sciq_pairs_samples_per_second": 22.088, + "eval_sciq_pairs_steps_per_second": 0.552, + "step": 9101 + }, + { + "epoch": 1.427383939774153, + "eval_qasc_pairs_loss": 0.4374461770057678, + "eval_qasc_pairs_runtime": 1.2131, + "eval_qasc_pairs_samples_per_second": 164.864, + "eval_qasc_pairs_steps_per_second": 4.122, + "step": 9101 + }, + { + "epoch": 1.427383939774153, + "eval_openbookqa_pairs_loss": 1.935410737991333, + "eval_openbookqa_pairs_runtime": 1.0551, + "eval_openbookqa_pairs_samples_per_second": 189.556, + "eval_openbookqa_pairs_steps_per_second": 4.739, + "step": 9101 + }, + { + "epoch": 1.427383939774153, + "eval_msmarco_pairs_loss": 1.043725609779358, + "eval_msmarco_pairs_runtime": 2.5259, + "eval_msmarco_pairs_samples_per_second": 79.179, + "eval_msmarco_pairs_steps_per_second": 1.979, + "step": 9101 + }, + { + "epoch": 1.427383939774153, + "eval_nq_pairs_loss": 0.9651347398757935, + "eval_nq_pairs_runtime": 5.6774, + "eval_nq_pairs_samples_per_second": 35.227, + "eval_nq_pairs_steps_per_second": 0.881, + "step": 9101 + }, + { + "epoch": 1.427383939774153, + "eval_trivia_pairs_loss": 1.2430963516235352, + "eval_trivia_pairs_runtime": 9.0564, + "eval_trivia_pairs_samples_per_second": 22.084, + "eval_trivia_pairs_steps_per_second": 0.552, + "step": 9101 + }, + { + "epoch": 1.427383939774153, + "eval_quora_pairs_loss": 0.21621538698673248, + "eval_quora_pairs_runtime": 0.6252, + "eval_quora_pairs_samples_per_second": 319.908, + "eval_quora_pairs_steps_per_second": 7.998, + "step": 9101 + }, + { + "epoch": 1.427383939774153, + "eval_gooaq_pairs_loss": 0.8927494287490845, + "eval_gooaq_pairs_runtime": 1.5775, + "eval_gooaq_pairs_samples_per_second": 126.779, + "eval_gooaq_pairs_steps_per_second": 3.169, + "step": 9101 + }, + { + "epoch": 1.427383939774153, + "eval_mrpc_pairs_loss": 0.08728782832622528, + "eval_mrpc_pairs_runtime": 0.2422, + "eval_mrpc_pairs_samples_per_second": 825.773, + "eval_mrpc_pairs_steps_per_second": 20.644, + "step": 9101 + }, + { + "epoch": 1.4303638644918444, + "grad_norm": 10.624916076660156, + "learning_rate": 1.1894318588256689e-05, + "loss": 1.2641, + "step": 9120 + }, + { + "epoch": 1.4378920953575909, + "grad_norm": 16.24142074584961, + "learning_rate": 1.1547069572392431e-05, + "loss": 1.4688, + "step": 9168 + }, + { + "epoch": 1.4454203262233376, + "grad_norm": 18.161853790283203, + "learning_rate": 1.120245140218616e-05, + "loss": 1.2061, + "step": 9216 + }, + { + "epoch": 1.452948557089084, + "grad_norm": 9.601255416870117, + "learning_rate": 1.0860616378631118e-05, + "loss": 1.0768, + "step": 9264 + }, + { + "epoch": 1.4604767879548306, + "grad_norm": 10.885648727416992, + "learning_rate": 1.0521715572733149e-05, + "loss": 0.9741, + "step": 9312 + }, + { + "epoch": 1.468005018820577, + "grad_norm": 4.2853007316589355, + "learning_rate": 1.018589875874622e-05, + "loss": 0.9977, + "step": 9360 + }, + { + "epoch": 1.4755332496863236, + "grad_norm": 14.765129089355469, + "learning_rate": 9.853314347980975e-06, + "loss": 1.308, + "step": 9408 + }, + { + "epoch": 1.4830614805520703, + "grad_norm": 0.6372986435890198, + "learning_rate": 9.524109323215664e-06, + "loss": 1.3859, + "step": 9456 + }, + { + "epoch": 1.4905897114178168, + "grad_norm": 5.864046573638916, + "learning_rate": 9.19842917373838e-06, + "loss": 0.9261, + "step": 9504 + }, + { + "epoch": 1.4981179422835633, + "grad_norm": 2.619722366333008, + "learning_rate": 8.87641783104935e-06, + "loss": 1.0473, + "step": 9552 + }, + { + "epoch": 1.502509410288582, + "eval_nli-pairs_loss": 1.0287954807281494, + "eval_nli-pairs_runtime": 4.1819, + "eval_nli-pairs_samples_per_second": 47.825, + "eval_nli-pairs_steps_per_second": 1.196, + "eval_sts-test_pearson_cosine": 0.772936269755451, + "eval_sts-test_pearson_dot": 0.5408568837341375, + "eval_sts-test_pearson_euclidean": 0.7431310400735589, + "eval_sts-test_pearson_manhattan": 0.7441695396815142, + "eval_sts-test_pearson_max": 0.772936269755451, + "eval_sts-test_spearman_cosine": 0.7808770614234607, + "eval_sts-test_spearman_dot": 0.5146216887404633, + "eval_sts-test_spearman_euclidean": 0.7291900713010305, + "eval_sts-test_spearman_manhattan": 0.732990460571053, + "eval_sts-test_spearman_max": 0.7808770614234607, + "step": 9580 + }, + { + "epoch": 1.502509410288582, + "eval_vitaminc-pairs_loss": 4.766480445861816, + "eval_vitaminc-pairs_runtime": 1.5156, + "eval_vitaminc-pairs_samples_per_second": 109.528, + "eval_vitaminc-pairs_steps_per_second": 2.639, + "step": 9580 + }, + { + "epoch": 1.502509410288582, + "eval_sts-label_loss": 4.150259494781494, + "eval_sts-label_runtime": 0.4493, + "eval_sts-label_samples_per_second": 445.153, + "eval_sts-label_steps_per_second": 11.129, + "step": 9580 + }, + { + "epoch": 1.502509410288582, + "eval_qnli-contrastive_loss": 0.1541556417942047, + "eval_qnli-contrastive_runtime": 0.296, + "eval_qnli-contrastive_samples_per_second": 675.668, + "eval_qnli-contrastive_steps_per_second": 16.892, + "step": 9580 + }, + { + "epoch": 1.502509410288582, + "eval_scitail-pairs-qa_loss": 0.1008567065000534, + "eval_scitail-pairs-qa_runtime": 1.0997, + "eval_scitail-pairs-qa_samples_per_second": 181.866, + "eval_scitail-pairs-qa_steps_per_second": 4.547, + "step": 9580 + }, + { + "epoch": 1.502509410288582, + "eval_scitail-pairs-pos_loss": 0.47411486506462097, + "eval_scitail-pairs-pos_runtime": 2.4599, + "eval_scitail-pairs-pos_samples_per_second": 81.304, + "eval_scitail-pairs-pos_steps_per_second": 2.033, + "step": 9580 + }, + { + "epoch": 1.502509410288582, + "eval_xsum-pairs_loss": 0.3825688064098358, + "eval_xsum-pairs_runtime": 1.0466, + "eval_xsum-pairs_samples_per_second": 191.088, + "eval_xsum-pairs_steps_per_second": 4.777, + "step": 9580 + }, + { + "epoch": 1.502509410288582, + "eval_compression-pairs_loss": 0.16390098631381989, + "eval_compression-pairs_runtime": 0.2542, + "eval_compression-pairs_samples_per_second": 786.663, + "eval_compression-pairs_steps_per_second": 19.667, + "step": 9580 + }, + { + "epoch": 1.502509410288582, + "eval_sciq_pairs_loss": 0.3327814042568207, + "eval_sciq_pairs_runtime": 9.1285, + "eval_sciq_pairs_samples_per_second": 21.909, + "eval_sciq_pairs_steps_per_second": 0.548, + "step": 9580 + }, + { + "epoch": 1.502509410288582, + "eval_qasc_pairs_loss": 0.42448753118515015, + "eval_qasc_pairs_runtime": 1.238, + "eval_qasc_pairs_samples_per_second": 161.554, + "eval_qasc_pairs_steps_per_second": 4.039, + "step": 9580 + }, + { + "epoch": 1.502509410288582, + "eval_openbookqa_pairs_loss": 1.9015610218048096, + "eval_openbookqa_pairs_runtime": 1.0684, + "eval_openbookqa_pairs_samples_per_second": 187.196, + "eval_openbookqa_pairs_steps_per_second": 4.68, + "step": 9580 + }, + { + "epoch": 1.502509410288582, + "eval_msmarco_pairs_loss": 1.049147605895996, + "eval_msmarco_pairs_runtime": 2.5566, + "eval_msmarco_pairs_samples_per_second": 78.23, + "eval_msmarco_pairs_steps_per_second": 1.956, + "step": 9580 + }, + { + "epoch": 1.502509410288582, + "eval_nq_pairs_loss": 0.938487708568573, + "eval_nq_pairs_runtime": 5.7298, + "eval_nq_pairs_samples_per_second": 34.905, + "eval_nq_pairs_steps_per_second": 0.873, + "step": 9580 + }, + { + "epoch": 1.502509410288582, + "eval_trivia_pairs_loss": 1.1729891300201416, + "eval_trivia_pairs_runtime": 9.1105, + "eval_trivia_pairs_samples_per_second": 21.953, + "eval_trivia_pairs_steps_per_second": 0.549, + "step": 9580 + }, + { + "epoch": 1.502509410288582, + "eval_quora_pairs_loss": 0.24502086639404297, + "eval_quora_pairs_runtime": 0.6539, + "eval_quora_pairs_samples_per_second": 305.842, + "eval_quora_pairs_steps_per_second": 7.646, + "step": 9580 + }, + { + "epoch": 1.502509410288582, + "eval_gooaq_pairs_loss": 0.8770759105682373, + "eval_gooaq_pairs_runtime": 1.6351, + "eval_gooaq_pairs_samples_per_second": 122.318, + "eval_gooaq_pairs_steps_per_second": 3.058, + "step": 9580 + }, + { + "epoch": 1.502509410288582, + "eval_mrpc_pairs_loss": 0.08623871207237244, + "eval_mrpc_pairs_runtime": 0.2404, + "eval_mrpc_pairs_samples_per_second": 832.077, + "eval_mrpc_pairs_steps_per_second": 20.802, + "step": 9580 + }, + { + "epoch": 1.50564617314931, + "grad_norm": 11.285247802734375, + "learning_rate": 8.558217605251645e-06, + "loss": 1.1806, + "step": 9600 + }, + { + "epoch": 1.5131744040150563, + "grad_norm": 10.830084800720215, + "learning_rate": 8.24396912215848e-06, + "loss": 1.1989, + "step": 9648 + }, + { + "epoch": 1.520702634880803, + "grad_norm": 9.796110153198242, + "learning_rate": 7.933811261144869e-06, + "loss": 1.0403, + "step": 9696 + }, + { + "epoch": 1.5282308657465495, + "grad_norm": 0.9950674176216125, + "learning_rate": 7.627881093771088e-06, + "loss": 1.0811, + "step": 9744 + }, + { + "epoch": 1.535759096612296, + "grad_norm": 11.353304862976074, + "learning_rate": 7.326313823205104e-06, + "loss": 1.3524, + "step": 9792 + }, + { + "epoch": 1.5432873274780428, + "grad_norm": 12.985937118530273, + "learning_rate": 7.029242724470705e-06, + "loss": 0.9578, + "step": 9840 + }, + { + "epoch": 1.5508155583437893, + "grad_norm": 8.885215759277344, + "learning_rate": 6.736799085547775e-06, + "loss": 1.2745, + "step": 9888 + }, + { + "epoch": 1.5583437892095358, + "grad_norm": 0.3098331391811371, + "learning_rate": 6.4491121493507095e-06, + "loss": 1.0615, + "step": 9936 + }, + { + "epoch": 1.5658720200752823, + "grad_norm": 14.668235778808594, + "learning_rate": 6.1663090566106425e-06, + "loss": 0.9778, + "step": 9984 + }, + { + "epoch": 1.5734002509410288, + "grad_norm": 14.634328842163086, + "learning_rate": 5.888514789686705e-06, + "loss": 1.017, + "step": 10032 + }, + { + "epoch": 1.5776348808030112, + "eval_nli-pairs_loss": 1.018608808517456, + "eval_nli-pairs_runtime": 3.9852, + "eval_nli-pairs_samples_per_second": 50.186, + "eval_nli-pairs_steps_per_second": 1.255, + "eval_sts-test_pearson_cosine": 0.7763527186202379, + "eval_sts-test_pearson_dot": 0.5518277089774203, + "eval_sts-test_pearson_euclidean": 0.7484152647439328, + "eval_sts-test_pearson_manhattan": 0.7486588645110376, + "eval_sts-test_pearson_max": 0.7763527186202379, + "eval_sts-test_spearman_cosine": 0.7862296389332735, + "eval_sts-test_spearman_dot": 0.5246898972209683, + "eval_sts-test_spearman_euclidean": 0.7340067918071436, + "eval_sts-test_spearman_manhattan": 0.7368940903638915, + "eval_sts-test_spearman_max": 0.7862296389332735, + "step": 10059 + }, + { + "epoch": 1.5776348808030112, + "eval_vitaminc-pairs_loss": 4.802022457122803, + "eval_vitaminc-pairs_runtime": 1.4418, + "eval_vitaminc-pairs_samples_per_second": 115.137, + "eval_vitaminc-pairs_steps_per_second": 2.774, + "step": 10059 + }, + { + "epoch": 1.5776348808030112, + "eval_sts-label_loss": 4.154313087463379, + "eval_sts-label_runtime": 0.4056, + "eval_sts-label_samples_per_second": 493.08, + "eval_sts-label_steps_per_second": 12.327, + "step": 10059 + }, + { + "epoch": 1.5776348808030112, + "eval_qnli-contrastive_loss": 0.1918843388557434, + "eval_qnli-contrastive_runtime": 0.2808, + "eval_qnli-contrastive_samples_per_second": 712.295, + "eval_qnli-contrastive_steps_per_second": 17.807, + "step": 10059 + }, + { + "epoch": 1.5776348808030112, + "eval_scitail-pairs-qa_loss": 0.10166393220424652, + "eval_scitail-pairs-qa_runtime": 1.0544, + "eval_scitail-pairs-qa_samples_per_second": 189.685, + "eval_scitail-pairs-qa_steps_per_second": 4.742, + "step": 10059 + }, + { + "epoch": 1.5776348808030112, + "eval_scitail-pairs-pos_loss": 0.4371533691883087, + "eval_scitail-pairs-pos_runtime": 2.4374, + "eval_scitail-pairs-pos_samples_per_second": 82.054, + "eval_scitail-pairs-pos_steps_per_second": 2.051, + "step": 10059 + }, + { + "epoch": 1.5776348808030112, + "eval_xsum-pairs_loss": 0.3739396035671234, + "eval_xsum-pairs_runtime": 1.0423, + "eval_xsum-pairs_samples_per_second": 191.881, + "eval_xsum-pairs_steps_per_second": 4.797, + "step": 10059 + }, + { + "epoch": 1.5776348808030112, + "eval_compression-pairs_loss": 0.16310901939868927, + "eval_compression-pairs_runtime": 0.2376, + "eval_compression-pairs_samples_per_second": 841.702, + "eval_compression-pairs_steps_per_second": 21.043, + "step": 10059 + }, + { + "epoch": 1.5776348808030112, + "eval_sciq_pairs_loss": 0.3224416971206665, + "eval_sciq_pairs_runtime": 9.0722, + "eval_sciq_pairs_samples_per_second": 22.045, + "eval_sciq_pairs_steps_per_second": 0.551, + "step": 10059 + }, + { + "epoch": 1.5776348808030112, + "eval_qasc_pairs_loss": 0.4264788031578064, + "eval_qasc_pairs_runtime": 1.2568, + "eval_qasc_pairs_samples_per_second": 159.131, + "eval_qasc_pairs_steps_per_second": 3.978, + "step": 10059 + }, + { + "epoch": 1.5776348808030112, + "eval_openbookqa_pairs_loss": 1.824275016784668, + "eval_openbookqa_pairs_runtime": 1.0491, + "eval_openbookqa_pairs_samples_per_second": 190.648, + "eval_openbookqa_pairs_steps_per_second": 4.766, + "step": 10059 + }, + { + "epoch": 1.5776348808030112, + "eval_msmarco_pairs_loss": 1.0266730785369873, + "eval_msmarco_pairs_runtime": 2.5174, + "eval_msmarco_pairs_samples_per_second": 79.447, + "eval_msmarco_pairs_steps_per_second": 1.986, + "step": 10059 + }, + { + "epoch": 1.5776348808030112, + "eval_nq_pairs_loss": 0.9473356604576111, + "eval_nq_pairs_runtime": 5.6496, + "eval_nq_pairs_samples_per_second": 35.401, + "eval_nq_pairs_steps_per_second": 0.885, + "step": 10059 + }, + { + "epoch": 1.5776348808030112, + "eval_trivia_pairs_loss": 1.104791522026062, + "eval_trivia_pairs_runtime": 9.073, + "eval_trivia_pairs_samples_per_second": 22.044, + "eval_trivia_pairs_steps_per_second": 0.551, + "step": 10059 + }, + { + "epoch": 1.5776348808030112, + "eval_quora_pairs_loss": 0.2634066939353943, + "eval_quora_pairs_runtime": 0.6209, + "eval_quora_pairs_samples_per_second": 322.127, + "eval_quora_pairs_steps_per_second": 8.053, + "step": 10059 + }, + { + "epoch": 1.5776348808030112, + "eval_gooaq_pairs_loss": 0.8539897203445435, + "eval_gooaq_pairs_runtime": 1.5537, + "eval_gooaq_pairs_samples_per_second": 128.725, + "eval_gooaq_pairs_steps_per_second": 3.218, + "step": 10059 + }, + { + "epoch": 1.5776348808030112, + "eval_mrpc_pairs_loss": 0.08671045303344727, + "eval_mrpc_pairs_runtime": 0.237, + "eval_mrpc_pairs_samples_per_second": 843.81, + "eval_mrpc_pairs_steps_per_second": 21.095, + "step": 10059 + }, + { + "epoch": 1.5809284818067755, + "grad_norm": 4.454440593719482, + "learning_rate": 5.615852117331175e-06, + "loss": 1.3413, + "step": 10080 + }, + { + "epoch": 1.588456712672522, + "grad_norm": 0.9061813950538635, + "learning_rate": 5.348441540432878e-06, + "loss": 1.1091, + "step": 10128 + }, + { + "epoch": 1.5959849435382685, + "grad_norm": 5.169127464294434, + "learning_rate": 5.086401238762887e-06, + "loss": 1.3101, + "step": 10176 + }, + { + "epoch": 1.6035131744040152, + "grad_norm": 4.494983196258545, + "learning_rate": 4.8298470187459895e-06, + "loss": 1.1126, + "step": 10224 + }, + { + "epoch": 1.6110414052697615, + "grad_norm": 0.4734310507774353, + "learning_rate": 4.578892262281069e-06, + "loss": 0.7667, + "step": 10272 + }, + { + "epoch": 1.6185696361355082, + "grad_norm": 0.4003587067127228, + "learning_rate": 4.333647876632947e-06, + "loss": 1.0281, + "step": 10320 + }, + { + "epoch": 1.6260978670012547, + "grad_norm": 1.0558795928955078, + "learning_rate": 4.094222245417886e-06, + "loss": 1.2234, + "step": 10368 + }, + { + "epoch": 1.6336260978670012, + "grad_norm": 1.4687271118164062, + "learning_rate": 3.860721180704409e-06, + "loss": 0.9402, + "step": 10416 + }, + { + "epoch": 1.641154328732748, + "grad_norm": 16.27718734741211, + "learning_rate": 3.6332478762505653e-06, + "loss": 0.9698, + "step": 10464 + }, + { + "epoch": 1.6486825595984942, + "grad_norm": 12.507079124450684, + "learning_rate": 3.411902861898354e-06, + "loss": 1.1373, + "step": 10512 + }, + { + "epoch": 1.6527603513174403, + "eval_nli-pairs_loss": 1.0013196468353271, + "eval_nli-pairs_runtime": 4.0021, + "eval_nli-pairs_samples_per_second": 49.974, + "eval_nli-pairs_steps_per_second": 1.249, + "eval_sts-test_pearson_cosine": 0.7756730383098025, + "eval_sts-test_pearson_dot": 0.547730552868467, + "eval_sts-test_pearson_euclidean": 0.7481152916902115, + "eval_sts-test_pearson_manhattan": 0.7486290345837316, + "eval_sts-test_pearson_max": 0.7756730383098025, + "eval_sts-test_spearman_cosine": 0.7844179279938138, + "eval_sts-test_spearman_dot": 0.5204811914426826, + "eval_sts-test_spearman_euclidean": 0.7335913694810947, + "eval_sts-test_spearman_manhattan": 0.7368797089125716, + "eval_sts-test_spearman_max": 0.7844179279938138, + "step": 10538 + }, + { + "epoch": 1.6527603513174403, + "eval_vitaminc-pairs_loss": 4.780285835266113, + "eval_vitaminc-pairs_runtime": 1.4992, + "eval_vitaminc-pairs_samples_per_second": 110.723, + "eval_vitaminc-pairs_steps_per_second": 2.668, + "step": 10538 + }, + { + "epoch": 1.6527603513174403, + "eval_sts-label_loss": 4.135310173034668, + "eval_sts-label_runtime": 0.4056, + "eval_sts-label_samples_per_second": 493.077, + "eval_sts-label_steps_per_second": 12.327, + "step": 10538 + }, + { + "epoch": 1.6527603513174403, + "eval_qnli-contrastive_loss": 0.18042831122875214, + "eval_qnli-contrastive_runtime": 0.2833, + "eval_qnli-contrastive_samples_per_second": 706.004, + "eval_qnli-contrastive_steps_per_second": 17.65, + "step": 10538 + }, + { + "epoch": 1.6527603513174403, + "eval_scitail-pairs-qa_loss": 0.09958843886852264, + "eval_scitail-pairs-qa_runtime": 1.0443, + "eval_scitail-pairs-qa_samples_per_second": 191.524, + "eval_scitail-pairs-qa_steps_per_second": 4.788, + "step": 10538 + }, + { + "epoch": 1.6527603513174403, + "eval_scitail-pairs-pos_loss": 0.4427280128002167, + "eval_scitail-pairs-pos_runtime": 2.3552, + "eval_scitail-pairs-pos_samples_per_second": 84.918, + "eval_scitail-pairs-pos_steps_per_second": 2.123, + "step": 10538 + }, + { + "epoch": 1.6527603513174403, + "eval_xsum-pairs_loss": 0.3688097298145294, + "eval_xsum-pairs_runtime": 1.041, + "eval_xsum-pairs_samples_per_second": 192.117, + "eval_xsum-pairs_steps_per_second": 4.803, + "step": 10538 + }, + { + "epoch": 1.6527603513174403, + "eval_compression-pairs_loss": 0.15796488523483276, + "eval_compression-pairs_runtime": 0.2407, + "eval_compression-pairs_samples_per_second": 830.774, + "eval_compression-pairs_steps_per_second": 20.769, + "step": 10538 + }, + { + "epoch": 1.6527603513174403, + "eval_sciq_pairs_loss": 0.3223775029182434, + "eval_sciq_pairs_runtime": 9.094, + "eval_sciq_pairs_samples_per_second": 21.992, + "eval_sciq_pairs_steps_per_second": 0.55, + "step": 10538 + }, + { + "epoch": 1.6527603513174403, + "eval_qasc_pairs_loss": 0.4077293872833252, + "eval_qasc_pairs_runtime": 1.2194, + "eval_qasc_pairs_samples_per_second": 164.013, + "eval_qasc_pairs_steps_per_second": 4.1, + "step": 10538 + }, + { + "epoch": 1.6527603513174403, + "eval_openbookqa_pairs_loss": 1.8293620347976685, + "eval_openbookqa_pairs_runtime": 1.0475, + "eval_openbookqa_pairs_samples_per_second": 190.939, + "eval_openbookqa_pairs_steps_per_second": 4.773, + "step": 10538 + }, + { + "epoch": 1.6527603513174403, + "eval_msmarco_pairs_loss": 1.0322593450546265, + "eval_msmarco_pairs_runtime": 2.5279, + "eval_msmarco_pairs_samples_per_second": 79.117, + "eval_msmarco_pairs_steps_per_second": 1.978, + "step": 10538 + }, + { + "epoch": 1.6527603513174403, + "eval_nq_pairs_loss": 0.9389599561691284, + "eval_nq_pairs_runtime": 5.6466, + "eval_nq_pairs_samples_per_second": 35.42, + "eval_nq_pairs_steps_per_second": 0.885, + "step": 10538 + }, + { + "epoch": 1.6527603513174403, + "eval_trivia_pairs_loss": 1.1042495965957642, + "eval_trivia_pairs_runtime": 9.0489, + "eval_trivia_pairs_samples_per_second": 22.102, + "eval_trivia_pairs_steps_per_second": 0.553, + "step": 10538 + }, + { + "epoch": 1.6527603513174403, + "eval_quora_pairs_loss": 0.17337936162948608, + "eval_quora_pairs_runtime": 0.6556, + "eval_quora_pairs_samples_per_second": 305.082, + "eval_quora_pairs_steps_per_second": 7.627, + "step": 10538 + }, + { + "epoch": 1.6527603513174403, + "eval_gooaq_pairs_loss": 0.8494808673858643, + "eval_gooaq_pairs_runtime": 1.5517, + "eval_gooaq_pairs_samples_per_second": 128.894, + "eval_gooaq_pairs_steps_per_second": 3.222, + "step": 10538 + }, + { + "epoch": 1.6527603513174403, + "eval_mrpc_pairs_loss": 0.08429060131311417, + "eval_mrpc_pairs_runtime": 0.2427, + "eval_mrpc_pairs_samples_per_second": 824.181, + "eval_mrpc_pairs_steps_per_second": 20.605, + "step": 10538 + }, + { + "epoch": 1.656210790464241, + "grad_norm": 9.579873085021973, + "learning_rate": 3.196783959145439e-06, + "loss": 1.433, + "step": 10560 + }, + { + "epoch": 1.6637390213299874, + "grad_norm": 20.199016571044922, + "learning_rate": 2.9879862379138003e-06, + "loss": 0.7482, + "step": 10608 + }, + { + "epoch": 1.671267252195734, + "grad_norm": 0.24710102379322052, + "learning_rate": 2.7897522915539458e-06, + "loss": 1.1968, + "step": 10656 + }, + { + "epoch": 1.6787954830614806, + "grad_norm": 9.337010383605957, + "learning_rate": 2.5937345579013605e-06, + "loss": 1.2424, + "step": 10704 + }, + { + "epoch": 1.6863237139272271, + "grad_norm": 13.432022094726562, + "learning_rate": 2.4043045181907913e-06, + "loss": 1.379, + "step": 10752 + }, + { + "epoch": 1.6938519447929736, + "grad_norm": 8.947489738464355, + "learning_rate": 2.2215458893753594e-06, + "loss": 1.4127, + "step": 10800 + }, + { + "epoch": 1.7013801756587204, + "grad_norm": 16.45635414123535, + "learning_rate": 2.045539440036334e-06, + "loss": 1.203, + "step": 10848 + }, + { + "epoch": 1.7089084065244666, + "grad_norm": 15.3703031539917, + "learning_rate": 1.8763629546881908e-06, + "loss": 0.7367, + "step": 10896 + }, + { + "epoch": 1.7164366373902133, + "grad_norm": 4.174514293670654, + "learning_rate": 1.714091199402421e-06, + "loss": 0.9207, + "step": 10944 + }, + { + "epoch": 1.7239648682559598, + "grad_norm": 2.0591108798980713, + "learning_rate": 1.558795888765306e-06, + "loss": 1.0067, + "step": 10992 + }, + { + "epoch": 1.7278858218318696, + "eval_nli-pairs_loss": 0.9888688325881958, + "eval_nli-pairs_runtime": 3.9929, + "eval_nli-pairs_samples_per_second": 50.089, + "eval_nli-pairs_steps_per_second": 1.252, + "eval_sts-test_pearson_cosine": 0.7760455578346633, + "eval_sts-test_pearson_dot": 0.5470527605127045, + "eval_sts-test_pearson_euclidean": 0.7462734713816128, + "eval_sts-test_pearson_manhattan": 0.7469300157882987, + "eval_sts-test_pearson_max": 0.7760455578346633, + "eval_sts-test_spearman_cosine": 0.7843247414472263, + "eval_sts-test_spearman_dot": 0.5195687627478929, + "eval_sts-test_spearman_euclidean": 0.7314459897716622, + "eval_sts-test_spearman_manhattan": 0.7349029838686436, + "eval_sts-test_spearman_max": 0.7843247414472263, + "step": 11017 + }, + { + "epoch": 1.7278858218318696, + "eval_vitaminc-pairs_loss": 4.805673122406006, + "eval_vitaminc-pairs_runtime": 1.4386, + "eval_vitaminc-pairs_samples_per_second": 115.388, + "eval_vitaminc-pairs_steps_per_second": 2.78, + "step": 11017 + }, + { + "epoch": 1.7278858218318696, + "eval_sts-label_loss": 4.089193344116211, + "eval_sts-label_runtime": 0.4049, + "eval_sts-label_samples_per_second": 493.975, + "eval_sts-label_steps_per_second": 12.349, + "step": 11017 + }, + { + "epoch": 1.7278858218318696, + "eval_qnli-contrastive_loss": 0.16640476882457733, + "eval_qnli-contrastive_runtime": 0.2871, + "eval_qnli-contrastive_samples_per_second": 696.519, + "eval_qnli-contrastive_steps_per_second": 17.413, + "step": 11017 + }, + { + "epoch": 1.7278858218318696, + "eval_scitail-pairs-qa_loss": 0.09589708596467972, + "eval_scitail-pairs-qa_runtime": 1.0968, + "eval_scitail-pairs-qa_samples_per_second": 182.341, + "eval_scitail-pairs-qa_steps_per_second": 4.559, + "step": 11017 + }, + { + "epoch": 1.7278858218318696, + "eval_scitail-pairs-pos_loss": 0.44937801361083984, + "eval_scitail-pairs-pos_runtime": 2.4011, + "eval_scitail-pairs-pos_samples_per_second": 83.296, + "eval_scitail-pairs-pos_steps_per_second": 2.082, + "step": 11017 + }, + { + "epoch": 1.7278858218318696, + "eval_xsum-pairs_loss": 0.3645179867744446, + "eval_xsum-pairs_runtime": 1.0406, + "eval_xsum-pairs_samples_per_second": 192.196, + "eval_xsum-pairs_steps_per_second": 4.805, + "step": 11017 + }, + { + "epoch": 1.7278858218318696, + "eval_compression-pairs_loss": 0.15408411622047424, + "eval_compression-pairs_runtime": 0.2368, + "eval_compression-pairs_samples_per_second": 844.756, + "eval_compression-pairs_steps_per_second": 21.119, + "step": 11017 + }, + { + "epoch": 1.7278858218318696, + "eval_sciq_pairs_loss": 0.3195297122001648, + "eval_sciq_pairs_runtime": 9.0709, + "eval_sciq_pairs_samples_per_second": 22.049, + "eval_sciq_pairs_steps_per_second": 0.551, + "step": 11017 + }, + { + "epoch": 1.7278858218318696, + "eval_qasc_pairs_loss": 0.40569430589675903, + "eval_qasc_pairs_runtime": 1.2166, + "eval_qasc_pairs_samples_per_second": 164.395, + "eval_qasc_pairs_steps_per_second": 4.11, + "step": 11017 + }, + { + "epoch": 1.7278858218318696, + "eval_openbookqa_pairs_loss": 1.8376811742782593, + "eval_openbookqa_pairs_runtime": 1.0469, + "eval_openbookqa_pairs_samples_per_second": 191.033, + "eval_openbookqa_pairs_steps_per_second": 4.776, + "step": 11017 + }, + { + "epoch": 1.7278858218318696, + "eval_msmarco_pairs_loss": 1.0271800756454468, + "eval_msmarco_pairs_runtime": 2.5127, + "eval_msmarco_pairs_samples_per_second": 79.597, + "eval_msmarco_pairs_steps_per_second": 1.99, + "step": 11017 + }, + { + "epoch": 1.7278858218318696, + "eval_nq_pairs_loss": 0.9162300229072571, + "eval_nq_pairs_runtime": 5.647, + "eval_nq_pairs_samples_per_second": 35.417, + "eval_nq_pairs_steps_per_second": 0.885, + "step": 11017 + }, + { + "epoch": 1.7278858218318696, + "eval_trivia_pairs_loss": 1.0903488397598267, + "eval_trivia_pairs_runtime": 9.0283, + "eval_trivia_pairs_samples_per_second": 22.153, + "eval_trivia_pairs_steps_per_second": 0.554, + "step": 11017 + }, + { + "epoch": 1.7278858218318696, + "eval_quora_pairs_loss": 0.19430270791053772, + "eval_quora_pairs_runtime": 0.6239, + "eval_quora_pairs_samples_per_second": 320.54, + "eval_quora_pairs_steps_per_second": 8.013, + "step": 11017 + }, + { + "epoch": 1.7278858218318696, + "eval_gooaq_pairs_loss": 0.8347020149230957, + "eval_gooaq_pairs_runtime": 1.5623, + "eval_gooaq_pairs_samples_per_second": 128.013, + "eval_gooaq_pairs_steps_per_second": 3.2, + "step": 11017 + }, + { + "epoch": 1.7278858218318696, + "eval_mrpc_pairs_loss": 0.08172078430652618, + "eval_mrpc_pairs_runtime": 0.2448, + "eval_mrpc_pairs_samples_per_second": 816.98, + "eval_mrpc_pairs_steps_per_second": 20.425, + "step": 11017 + }, + { + "epoch": 1.7314930991217063, + "grad_norm": 16.512481689453125, + "learning_rate": 1.410545654184303e-06, + "loss": 0.8782, + "step": 11040 + }, + { + "epoch": 1.739021329987453, + "grad_norm": 15.815868377685547, + "learning_rate": 1.2694060135569684e-06, + "loss": 1.1617, + "step": 11088 + }, + { + "epoch": 1.7465495608531993, + "grad_norm": 0.39903897047042847, + "learning_rate": 1.135439342315913e-06, + "loss": 1.0122, + "step": 11136 + }, + { + "epoch": 1.754077791718946, + "grad_norm": 7.398995876312256, + "learning_rate": 1.0087048458625138e-06, + "loss": 0.6694, + "step": 11184 + }, + { + "epoch": 1.7616060225846926, + "grad_norm": 10.535834312438965, + "learning_rate": 8.892585334016114e-07, + "loss": 1.1209, + "step": 11232 + }, + { + "epoch": 1.769134253450439, + "grad_norm": 14.994894027709961, + "learning_rate": 7.771531931887249e-07, + "loss": 1.1, + "step": 11280 + }, + { + "epoch": 1.7766624843161858, + "grad_norm": 29.789087295532227, + "learning_rate": 6.724383692007457e-07, + "loss": 0.9796, + "step": 11328 + }, + { + "epoch": 1.7841907151819323, + "grad_norm": 16.97585105895996, + "learning_rate": 5.751603392404059e-07, + "loss": 0.9533, + "step": 11376 + }, + { + "epoch": 1.7917189460476788, + "grad_norm": 36.554412841796875, + "learning_rate": 4.853620944842135e-07, + "loss": 0.9207, + "step": 11424 + }, + { + "epoch": 1.7992471769134255, + "grad_norm": 12.20507526397705, + "learning_rate": 4.030833204828687e-07, + "loss": 1.0267, + "step": 11472 + }, + { + "epoch": 1.8030112923462986, + "eval_nli-pairs_loss": 0.9834117293357849, + "eval_nli-pairs_runtime": 4.454, + "eval_nli-pairs_samples_per_second": 44.904, + "eval_nli-pairs_steps_per_second": 1.123, + "eval_sts-test_pearson_cosine": 0.7757193524880511, + "eval_sts-test_pearson_dot": 0.5461517523115149, + "eval_sts-test_pearson_euclidean": 0.7464325603201095, + "eval_sts-test_pearson_manhattan": 0.7470492681562052, + "eval_sts-test_pearson_max": 0.7757193524880511, + "eval_sts-test_spearman_cosine": 0.7841895178722113, + "eval_sts-test_spearman_dot": 0.5190518591730641, + "eval_sts-test_spearman_euclidean": 0.7317065051907755, + "eval_sts-test_spearman_manhattan": 0.7348866670601053, + "eval_sts-test_spearman_max": 0.7841895178722113, + "step": 11496 + }, + { + "epoch": 1.8030112923462986, + "eval_vitaminc-pairs_loss": 4.804275989532471, + "eval_vitaminc-pairs_runtime": 1.4573, + "eval_vitaminc-pairs_samples_per_second": 113.908, + "eval_vitaminc-pairs_steps_per_second": 2.745, + "step": 11496 + }, + { + "epoch": 1.8030112923462986, + "eval_sts-label_loss": 4.131558418273926, + "eval_sts-label_runtime": 0.4133, + "eval_sts-label_samples_per_second": 483.878, + "eval_sts-label_steps_per_second": 12.097, + "step": 11496 + }, + { + "epoch": 1.8030112923462986, + "eval_qnli-contrastive_loss": 0.16553626954555511, + "eval_qnli-contrastive_runtime": 0.2817, + "eval_qnli-contrastive_samples_per_second": 709.973, + "eval_qnli-contrastive_steps_per_second": 17.749, + "step": 11496 + }, + { + "epoch": 1.8030112923462986, + "eval_scitail-pairs-qa_loss": 0.09527866542339325, + "eval_scitail-pairs-qa_runtime": 1.0708, + "eval_scitail-pairs-qa_samples_per_second": 186.771, + "eval_scitail-pairs-qa_steps_per_second": 4.669, + "step": 11496 + }, + { + "epoch": 1.8030112923462986, + "eval_scitail-pairs-pos_loss": 0.446532279253006, + "eval_scitail-pairs-pos_runtime": 2.3976, + "eval_scitail-pairs-pos_samples_per_second": 83.418, + "eval_scitail-pairs-pos_steps_per_second": 2.085, + "step": 11496 + }, + { + "epoch": 1.8030112923462986, + "eval_xsum-pairs_loss": 0.36277005076408386, + "eval_xsum-pairs_runtime": 1.0497, + "eval_xsum-pairs_samples_per_second": 190.534, + "eval_xsum-pairs_steps_per_second": 4.763, + "step": 11496 + }, + { + "epoch": 1.8030112923462986, + "eval_compression-pairs_loss": 0.15356417000293732, + "eval_compression-pairs_runtime": 0.2565, + "eval_compression-pairs_samples_per_second": 779.676, + "eval_compression-pairs_steps_per_second": 19.492, + "step": 11496 + }, + { + "epoch": 1.8030112923462986, + "eval_sciq_pairs_loss": 0.3197508752346039, + "eval_sciq_pairs_runtime": 9.2253, + "eval_sciq_pairs_samples_per_second": 21.679, + "eval_sciq_pairs_steps_per_second": 0.542, + "step": 11496 + }, + { + "epoch": 1.8030112923462986, + "eval_qasc_pairs_loss": 0.40177223086357117, + "eval_qasc_pairs_runtime": 1.2313, + "eval_qasc_pairs_samples_per_second": 162.427, + "eval_qasc_pairs_steps_per_second": 4.061, + "step": 11496 + }, + { + "epoch": 1.8030112923462986, + "eval_openbookqa_pairs_loss": 1.8343558311462402, + "eval_openbookqa_pairs_runtime": 1.0933, + "eval_openbookqa_pairs_samples_per_second": 182.939, + "eval_openbookqa_pairs_steps_per_second": 4.573, + "step": 11496 + }, + { + "epoch": 1.8030112923462986, + "eval_msmarco_pairs_loss": 1.0230004787445068, + "eval_msmarco_pairs_runtime": 2.5513, + "eval_msmarco_pairs_samples_per_second": 78.392, + "eval_msmarco_pairs_steps_per_second": 1.96, + "step": 11496 + }, + { + "epoch": 1.8030112923462986, + "eval_nq_pairs_loss": 0.9057186841964722, + "eval_nq_pairs_runtime": 5.7628, + "eval_nq_pairs_samples_per_second": 34.705, + "eval_nq_pairs_steps_per_second": 0.868, + "step": 11496 + }, + { + "epoch": 1.8030112923462986, + "eval_trivia_pairs_loss": 1.0895284414291382, + "eval_trivia_pairs_runtime": 9.1178, + "eval_trivia_pairs_samples_per_second": 21.935, + "eval_trivia_pairs_steps_per_second": 0.548, + "step": 11496 + }, + { + "epoch": 1.8030112923462986, + "eval_quora_pairs_loss": 0.1568813920021057, + "eval_quora_pairs_runtime": 0.6394, + "eval_quora_pairs_samples_per_second": 312.772, + "eval_quora_pairs_steps_per_second": 7.819, + "step": 11496 + }, + { + "epoch": 1.8030112923462986, + "eval_gooaq_pairs_loss": 0.8383786082267761, + "eval_gooaq_pairs_runtime": 1.5566, + "eval_gooaq_pairs_samples_per_second": 128.488, + "eval_gooaq_pairs_steps_per_second": 3.212, + "step": 11496 + }, + { + "epoch": 1.8030112923462986, + "eval_mrpc_pairs_loss": 0.08112096786499023, + "eval_mrpc_pairs_runtime": 0.2439, + "eval_mrpc_pairs_samples_per_second": 819.882, + "eval_mrpc_pairs_steps_per_second": 20.497, + "step": 11496 + }, + { + "epoch": 1.8067754077791718, + "grad_norm": 24.458271026611328, + "learning_rate": 3.2836037962258505e-07, + "loss": 0.8635, + "step": 11520 + }, + { + "epoch": 1.8143036386449185, + "grad_norm": 6.1133222579956055, + "learning_rate": 2.612262950550589e-07, + "loss": 0.9252, + "step": 11568 + }, + { + "epoch": 1.821831869510665, + "grad_norm": 10.278335571289062, + "learning_rate": 2.0171073610316003e-07, + "loss": 1.32, + "step": 11616 + }, + { + "epoch": 1.8293601003764115, + "grad_norm": 14.742138862609863, + "learning_rate": 1.4984000514884242e-07, + "loss": 1.3209, + "step": 11664 + }, + { + "epoch": 1.8368883312421582, + "grad_norm": 7.676764011383057, + "learning_rate": 1.056370260090303e-07, + "loss": 0.8251, + "step": 11712 + }, + { + "epoch": 1.8444165621079045, + "grad_norm": 1.2442536354064941, + "learning_rate": 6.912133380464951e-08, + "loss": 0.8179, + "step": 11760 + }, + { + "epoch": 1.8519447929736512, + "grad_norm": 15.847275733947754, + "learning_rate": 4.030906632723735e-08, + "loss": 0.8666, + "step": 11808 + }, + { + "epoch": 1.8594730238393977, + "grad_norm": 0.36177679896354675, + "learning_rate": 1.9212956906992372e-08, + "loss": 0.9244, + "step": 11856 + }, + { + "epoch": 1.8670012547051442, + "grad_norm": 8.72006607055664, + "learning_rate": 5.842328785392869e-09, + "loss": 0.9468, + "step": 11904 + }, + { + "epoch": 1.874529485570891, + "grad_norm": 15.35616397857666, + "learning_rate": 2.0309099486470748e-10, + "loss": 1.2313, + "step": 11952 + }, + { + "epoch": 1.8781367628607277, + "eval_nli-pairs_loss": 1.0213488340377808, + "eval_nli-pairs_runtime": 3.9888, + "eval_nli-pairs_samples_per_second": 50.14, + "eval_nli-pairs_steps_per_second": 1.254, + "eval_sts-test_pearson_cosine": 0.7751861574983592, + "eval_sts-test_pearson_dot": 0.5406109395741264, + "eval_sts-test_pearson_euclidean": 0.7432969762219094, + "eval_sts-test_pearson_manhattan": 0.7450333218883929, + "eval_sts-test_pearson_max": 0.7751861574983592, + "eval_sts-test_spearman_cosine": 0.7794274058735736, + "eval_sts-test_spearman_dot": 0.5144540377132975, + "eval_sts-test_spearman_euclidean": 0.727450852531023, + "eval_sts-test_spearman_manhattan": 0.7318298848990707, + "eval_sts-test_spearman_max": 0.7794274058735736, + "step": 11975 + }, + { + "epoch": 1.8781367628607277, + "eval_vitaminc-pairs_loss": 4.819454193115234, + "eval_vitaminc-pairs_runtime": 1.4385, + "eval_vitaminc-pairs_samples_per_second": 115.395, + "eval_vitaminc-pairs_steps_per_second": 2.781, + "step": 11975 + }, + { + "epoch": 1.8781367628607277, + "eval_sts-label_loss": 4.0922932624816895, + "eval_sts-label_runtime": 0.403, + "eval_sts-label_samples_per_second": 496.282, + "eval_sts-label_steps_per_second": 12.407, + "step": 11975 + }, + { + "epoch": 1.8781367628607277, + "eval_qnli-contrastive_loss": 0.14450308680534363, + "eval_qnli-contrastive_runtime": 0.2805, + "eval_qnli-contrastive_samples_per_second": 712.965, + "eval_qnli-contrastive_steps_per_second": 17.824, + "step": 11975 + }, + { + "epoch": 1.8781367628607277, + "eval_scitail-pairs-qa_loss": 0.0972411260008812, + "eval_scitail-pairs-qa_runtime": 1.0535, + "eval_scitail-pairs-qa_samples_per_second": 189.845, + "eval_scitail-pairs-qa_steps_per_second": 4.746, + "step": 11975 + }, + { + "epoch": 1.8781367628607277, + "eval_scitail-pairs-pos_loss": 0.47073617577552795, + "eval_scitail-pairs-pos_runtime": 2.4056, + "eval_scitail-pairs-pos_samples_per_second": 83.141, + "eval_scitail-pairs-pos_steps_per_second": 2.079, + "step": 11975 + }, + { + "epoch": 1.8781367628607277, + "eval_xsum-pairs_loss": 0.3865247964859009, + "eval_xsum-pairs_runtime": 1.0467, + "eval_xsum-pairs_samples_per_second": 191.083, + "eval_xsum-pairs_steps_per_second": 4.777, + "step": 11975 + }, + { + "epoch": 1.8781367628607277, + "eval_compression-pairs_loss": 0.15423807501792908, + "eval_compression-pairs_runtime": 0.2378, + "eval_compression-pairs_samples_per_second": 841.003, + "eval_compression-pairs_steps_per_second": 21.025, + "step": 11975 + }, + { + "epoch": 1.8781367628607277, + "eval_sciq_pairs_loss": 0.3360276520252228, + "eval_sciq_pairs_runtime": 9.0552, + "eval_sciq_pairs_samples_per_second": 22.087, + "eval_sciq_pairs_steps_per_second": 0.552, + "step": 11975 + }, + { + "epoch": 1.8781367628607277, + "eval_qasc_pairs_loss": 0.40982764959335327, + "eval_qasc_pairs_runtime": 1.2436, + "eval_qasc_pairs_samples_per_second": 160.828, + "eval_qasc_pairs_steps_per_second": 4.021, + "step": 11975 + }, + { + "epoch": 1.8781367628607277, + "eval_openbookqa_pairs_loss": 1.8729889392852783, + "eval_openbookqa_pairs_runtime": 1.106, + "eval_openbookqa_pairs_samples_per_second": 180.839, + "eval_openbookqa_pairs_steps_per_second": 4.521, + "step": 11975 + }, + { + "epoch": 1.8781367628607277, + "eval_msmarco_pairs_loss": 1.0765292644500732, + "eval_msmarco_pairs_runtime": 2.5194, + "eval_msmarco_pairs_samples_per_second": 79.385, + "eval_msmarco_pairs_steps_per_second": 1.985, + "step": 11975 + }, + { + "epoch": 1.8781367628607277, + "eval_nq_pairs_loss": 0.9723155498504639, + "eval_nq_pairs_runtime": 5.6285, + "eval_nq_pairs_samples_per_second": 35.534, + "eval_nq_pairs_steps_per_second": 0.888, + "step": 11975 + }, + { + "epoch": 1.8781367628607277, + "eval_trivia_pairs_loss": 1.2097830772399902, + "eval_trivia_pairs_runtime": 9.0412, + "eval_trivia_pairs_samples_per_second": 22.121, + "eval_trivia_pairs_steps_per_second": 0.553, + "step": 11975 + }, + { + "epoch": 1.8781367628607277, + "eval_quora_pairs_loss": 0.1777983158826828, + "eval_quora_pairs_runtime": 0.6175, + "eval_quora_pairs_samples_per_second": 323.895, + "eval_quora_pairs_steps_per_second": 8.097, + "step": 11975 + }, + { + "epoch": 1.8781367628607277, + "eval_gooaq_pairs_loss": 0.8855485320091248, + "eval_gooaq_pairs_runtime": 1.5417, + "eval_gooaq_pairs_samples_per_second": 129.724, + "eval_gooaq_pairs_steps_per_second": 3.243, + "step": 11975 + }, + { + "epoch": 1.8781367628607277, + "eval_mrpc_pairs_loss": 0.0837549939751625, + "eval_mrpc_pairs_runtime": 0.2368, + "eval_mrpc_pairs_samples_per_second": 844.455, + "eval_mrpc_pairs_steps_per_second": 21.111, + "step": 11975 + }, + { + "epoch": 1.8820577164366374, + "grad_norm": 6.167259216308594, + "learning_rate": 3.4997702264252654e-05, + "loss": 1.0019, + "step": 12000 + }, + { + "epoch": 1.889585947302384, + "grad_norm": 0.49681809544563293, + "learning_rate": 3.498787466266714e-05, + "loss": 0.8415, + "step": 12048 + }, + { + "epoch": 1.8971141781681304, + "grad_norm": 11.792595863342285, + "learning_rate": 3.4970318447471354e-05, + "loss": 1.0103, + "step": 12096 + }, + { + "epoch": 1.904642409033877, + "grad_norm": 0.8563029766082764, + "learning_rate": 3.494504137748141e-05, + "loss": 1.1249, + "step": 12144 + }, + { + "epoch": 1.9121706398996237, + "grad_norm": 36.67128372192383, + "learning_rate": 3.4912054623677835e-05, + "loss": 1.2655, + "step": 12192 + }, + { + "epoch": 1.9196988707653702, + "grad_norm": 15.591288566589355, + "learning_rate": 3.4871372764268616e-05, + "loss": 0.8125, + "step": 12240 + }, + { + "epoch": 1.9272271016311167, + "grad_norm": 0.8200180530548096, + "learning_rate": 3.482301377824655e-05, + "loss": 0.8559, + "step": 12288 + }, + { + "epoch": 1.9347553324968634, + "grad_norm": 0.6030488014221191, + "learning_rate": 3.476699903744353e-05, + "loss": 0.8416, + "step": 12336 + }, + { + "epoch": 1.9422835633626097, + "grad_norm": 17.357330322265625, + "learning_rate": 3.470335329708547e-05, + "loss": 1.0393, + "step": 12384 + }, + { + "epoch": 1.9498117942283564, + "grad_norm": 15.783160209655762, + "learning_rate": 3.463210468485197e-05, + "loss": 0.946, + "step": 12432 + }, + { + "epoch": 1.9532622333751568, + "eval_nli-pairs_loss": 1.0603748559951782, + "eval_nli-pairs_runtime": 4.0941, + "eval_nli-pairs_samples_per_second": 48.85, + "eval_nli-pairs_steps_per_second": 1.221, + "eval_sts-test_pearson_cosine": 0.7721603126678054, + "eval_sts-test_pearson_dot": 0.5349835988482088, + "eval_sts-test_pearson_euclidean": 0.742188046420877, + "eval_sts-test_pearson_manhattan": 0.740692747387156, + "eval_sts-test_pearson_max": 0.7721603126678054, + "eval_sts-test_spearman_cosine": 0.7864643989345994, + "eval_sts-test_spearman_dot": 0.5098554969670107, + "eval_sts-test_spearman_euclidean": 0.7318205135578197, + "eval_sts-test_spearman_manhattan": 0.7339408946246045, + "eval_sts-test_spearman_max": 0.7864643989345994, + "step": 12454 + }, + { + "epoch": 1.9532622333751568, + "eval_vitaminc-pairs_loss": 4.802134990692139, + "eval_vitaminc-pairs_runtime": 1.4328, + "eval_vitaminc-pairs_samples_per_second": 115.859, + "eval_vitaminc-pairs_steps_per_second": 2.792, + "step": 12454 + }, + { + "epoch": 1.9532622333751568, + "eval_sts-label_loss": 4.164713382720947, + "eval_sts-label_runtime": 0.4062, + "eval_sts-label_samples_per_second": 492.426, + "eval_sts-label_steps_per_second": 12.311, + "step": 12454 + }, + { + "epoch": 1.9532622333751568, + "eval_qnli-contrastive_loss": 0.1906559020280838, + "eval_qnli-contrastive_runtime": 0.285, + "eval_qnli-contrastive_samples_per_second": 701.782, + "eval_qnli-contrastive_steps_per_second": 17.545, + "step": 12454 + }, + { + "epoch": 1.9532622333751568, + "eval_scitail-pairs-qa_loss": 0.09206719696521759, + "eval_scitail-pairs-qa_runtime": 1.0605, + "eval_scitail-pairs-qa_samples_per_second": 188.59, + "eval_scitail-pairs-qa_steps_per_second": 4.715, + "step": 12454 + }, + { + "epoch": 1.9532622333751568, + "eval_scitail-pairs-pos_loss": 0.4270685613155365, + "eval_scitail-pairs-pos_runtime": 2.3609, + "eval_scitail-pairs-pos_samples_per_second": 84.714, + "eval_scitail-pairs-pos_steps_per_second": 2.118, + "step": 12454 + }, + { + "epoch": 1.9532622333751568, + "eval_xsum-pairs_loss": 0.386574923992157, + "eval_xsum-pairs_runtime": 1.0761, + "eval_xsum-pairs_samples_per_second": 185.848, + "eval_xsum-pairs_steps_per_second": 4.646, + "step": 12454 + }, + { + "epoch": 1.9532622333751568, + "eval_compression-pairs_loss": 0.15844617784023285, + "eval_compression-pairs_runtime": 0.2411, + "eval_compression-pairs_samples_per_second": 829.616, + "eval_compression-pairs_steps_per_second": 20.74, + "step": 12454 + }, + { + "epoch": 1.9532622333751568, + "eval_sciq_pairs_loss": 0.32742640376091003, + "eval_sciq_pairs_runtime": 9.0687, + "eval_sciq_pairs_samples_per_second": 22.054, + "eval_sciq_pairs_steps_per_second": 0.551, + "step": 12454 + }, + { + "epoch": 1.9532622333751568, + "eval_qasc_pairs_loss": 0.44627976417541504, + "eval_qasc_pairs_runtime": 1.2115, + "eval_qasc_pairs_samples_per_second": 165.083, + "eval_qasc_pairs_steps_per_second": 4.127, + "step": 12454 + }, + { + "epoch": 1.9532622333751568, + "eval_openbookqa_pairs_loss": 1.8450264930725098, + "eval_openbookqa_pairs_runtime": 1.0468, + "eval_openbookqa_pairs_samples_per_second": 191.053, + "eval_openbookqa_pairs_steps_per_second": 4.776, + "step": 12454 + }, + { + "epoch": 1.9532622333751568, + "eval_msmarco_pairs_loss": 1.0895458459854126, + "eval_msmarco_pairs_runtime": 2.5207, + "eval_msmarco_pairs_samples_per_second": 79.343, + "eval_msmarco_pairs_steps_per_second": 1.984, + "step": 12454 + }, + { + "epoch": 1.9532622333751568, + "eval_nq_pairs_loss": 0.9783583283424377, + "eval_nq_pairs_runtime": 5.6317, + "eval_nq_pairs_samples_per_second": 35.513, + "eval_nq_pairs_steps_per_second": 0.888, + "step": 12454 + }, + { + "epoch": 1.9532622333751568, + "eval_trivia_pairs_loss": 1.1956011056900024, + "eval_trivia_pairs_runtime": 9.0547, + "eval_trivia_pairs_samples_per_second": 22.088, + "eval_trivia_pairs_steps_per_second": 0.552, + "step": 12454 + }, + { + "epoch": 1.9532622333751568, + "eval_quora_pairs_loss": 0.2143821269273758, + "eval_quora_pairs_runtime": 0.6206, + "eval_quora_pairs_samples_per_second": 322.293, + "eval_quora_pairs_steps_per_second": 8.057, + "step": 12454 + }, + { + "epoch": 1.9532622333751568, + "eval_gooaq_pairs_loss": 0.890216052532196, + "eval_gooaq_pairs_runtime": 1.545, + "eval_gooaq_pairs_samples_per_second": 129.45, + "eval_gooaq_pairs_steps_per_second": 3.236, + "step": 12454 + }, + { + "epoch": 1.9532622333751568, + "eval_mrpc_pairs_loss": 0.08200729638338089, + "eval_mrpc_pairs_runtime": 0.2402, + "eval_mrpc_pairs_samples_per_second": 832.672, + "eval_mrpc_pairs_steps_per_second": 20.817, + "step": 12454 + }, + { + "epoch": 1.9573400250941029, + "grad_norm": 9.203104972839355, + "learning_rate": 3.455328468844549e-05, + "loss": 1.0748, + "step": 12480 + }, + { + "epoch": 1.9648682559598494, + "grad_norm": 1.351884126663208, + "learning_rate": 3.4466928141675676e-05, + "loss": 0.8413, + "step": 12528 + }, + { + "epoch": 1.972396486825596, + "grad_norm": 13.26588249206543, + "learning_rate": 3.4375104727660065e-05, + "loss": 1.2138, + "step": 12576 + }, + { + "epoch": 1.9799247176913424, + "grad_norm": 7.214226245880127, + "learning_rate": 3.427394779454246e-05, + "loss": 1.2554, + "step": 12624 + }, + { + "epoch": 1.987452948557089, + "grad_norm": 9.377580642700195, + "learning_rate": 3.416537776156432e-05, + "loss": 1.4695, + "step": 12672 + }, + { + "epoch": 1.9949811794228356, + "grad_norm": 14.431234359741211, + "learning_rate": 3.4049442610304357e-05, + "loss": 0.9206, + "step": 12720 + }, + { + "epoch": 2.002509410288582, + "grad_norm": 39.02935028076172, + "learning_rate": 3.392619357729103e-05, + "loss": 0.9354, + "step": 12768 + }, + { + "epoch": 2.010037641154329, + "grad_norm": 0.6344715356826782, + "learning_rate": 3.379568513135902e-05, + "loss": 0.8492, + "step": 12816 + }, + { + "epoch": 2.017565872020075, + "grad_norm": 12.117080688476562, + "learning_rate": 3.365797494957723e-05, + "loss": 0.9937, + "step": 12864 + }, + { + "epoch": 2.025094102885822, + "grad_norm": 8.92581844329834, + "learning_rate": 3.351312389175882e-05, + "loss": 1.423, + "step": 12912 + }, + { + "epoch": 2.028387703889586, + "eval_nli-pairs_loss": 1.1507288217544556, + "eval_nli-pairs_runtime": 3.9823, + "eval_nli-pairs_samples_per_second": 50.222, + "eval_nli-pairs_steps_per_second": 1.256, + "eval_sts-test_pearson_cosine": 0.7722624362390709, + "eval_sts-test_pearson_dot": 0.5303764984983903, + "eval_sts-test_pearson_euclidean": 0.7411643880467264, + "eval_sts-test_pearson_manhattan": 0.7391579789846456, + "eval_sts-test_pearson_max": 0.7722624362390709, + "eval_sts-test_spearman_cosine": 0.7821947805835431, + "eval_sts-test_spearman_dot": 0.5068928313854526, + "eval_sts-test_spearman_euclidean": 0.7301088773642002, + "eval_sts-test_spearman_manhattan": 0.7326196614978934, + "eval_sts-test_spearman_max": 0.7821947805835431, + "step": 12933 + }, + { + "epoch": 2.028387703889586, + "eval_vitaminc-pairs_loss": 4.279026985168457, + "eval_vitaminc-pairs_runtime": 1.4397, + "eval_vitaminc-pairs_samples_per_second": 115.303, + "eval_vitaminc-pairs_steps_per_second": 2.778, + "step": 12933 + }, + { + "epoch": 2.028387703889586, + "eval_sts-label_loss": 4.452165603637695, + "eval_sts-label_runtime": 0.4088, + "eval_sts-label_samples_per_second": 489.236, + "eval_sts-label_steps_per_second": 12.231, + "step": 12933 + }, + { + "epoch": 2.028387703889586, + "eval_qnli-contrastive_loss": 0.17014659941196442, + "eval_qnli-contrastive_runtime": 0.2882, + "eval_qnli-contrastive_samples_per_second": 693.965, + "eval_qnli-contrastive_steps_per_second": 17.349, + "step": 12933 + }, + { + "epoch": 2.028387703889586, + "eval_scitail-pairs-qa_loss": 0.09296510368585587, + "eval_scitail-pairs-qa_runtime": 1.0809, + "eval_scitail-pairs-qa_samples_per_second": 185.023, + "eval_scitail-pairs-qa_steps_per_second": 4.626, + "step": 12933 + }, + { + "epoch": 2.028387703889586, + "eval_scitail-pairs-pos_loss": 0.4562944173812866, + "eval_scitail-pairs-pos_runtime": 2.376, + "eval_scitail-pairs-pos_samples_per_second": 84.175, + "eval_scitail-pairs-pos_steps_per_second": 2.104, + "step": 12933 + }, + { + "epoch": 2.028387703889586, + "eval_xsum-pairs_loss": 0.3622417151927948, + "eval_xsum-pairs_runtime": 1.0416, + "eval_xsum-pairs_samples_per_second": 192.012, + "eval_xsum-pairs_steps_per_second": 4.8, + "step": 12933 + }, + { + "epoch": 2.028387703889586, + "eval_compression-pairs_loss": 0.15833701193332672, + "eval_compression-pairs_runtime": 0.2407, + "eval_compression-pairs_samples_per_second": 831.007, + "eval_compression-pairs_steps_per_second": 20.775, + "step": 12933 + }, + { + "epoch": 2.028387703889586, + "eval_sciq_pairs_loss": 0.3300960958003998, + "eval_sciq_pairs_runtime": 9.1169, + "eval_sciq_pairs_samples_per_second": 21.937, + "eval_sciq_pairs_steps_per_second": 0.548, + "step": 12933 + }, + { + "epoch": 2.028387703889586, + "eval_qasc_pairs_loss": 0.4440248906612396, + "eval_qasc_pairs_runtime": 1.2186, + "eval_qasc_pairs_samples_per_second": 164.116, + "eval_qasc_pairs_steps_per_second": 4.103, + "step": 12933 + }, + { + "epoch": 2.028387703889586, + "eval_openbookqa_pairs_loss": 1.9062250852584839, + "eval_openbookqa_pairs_runtime": 1.054, + "eval_openbookqa_pairs_samples_per_second": 189.748, + "eval_openbookqa_pairs_steps_per_second": 4.744, + "step": 12933 + }, + { + "epoch": 2.028387703889586, + "eval_msmarco_pairs_loss": 1.0610954761505127, + "eval_msmarco_pairs_runtime": 2.5303, + "eval_msmarco_pairs_samples_per_second": 79.042, + "eval_msmarco_pairs_steps_per_second": 1.976, + "step": 12933 + }, + { + "epoch": 2.028387703889586, + "eval_nq_pairs_loss": 0.9248062968254089, + "eval_nq_pairs_runtime": 5.6477, + "eval_nq_pairs_samples_per_second": 35.413, + "eval_nq_pairs_steps_per_second": 0.885, + "step": 12933 + }, + { + "epoch": 2.028387703889586, + "eval_trivia_pairs_loss": 1.1339099407196045, + "eval_trivia_pairs_runtime": 9.046, + "eval_trivia_pairs_samples_per_second": 22.109, + "eval_trivia_pairs_steps_per_second": 0.553, + "step": 12933 + }, + { + "epoch": 2.028387703889586, + "eval_quora_pairs_loss": 0.197435200214386, + "eval_quora_pairs_runtime": 0.6181, + "eval_quora_pairs_samples_per_second": 323.589, + "eval_quora_pairs_steps_per_second": 8.09, + "step": 12933 + }, + { + "epoch": 2.028387703889586, + "eval_gooaq_pairs_loss": 0.8682229518890381, + "eval_gooaq_pairs_runtime": 1.5695, + "eval_gooaq_pairs_samples_per_second": 127.427, + "eval_gooaq_pairs_steps_per_second": 3.186, + "step": 12933 + }, + { + "epoch": 2.028387703889586, + "eval_mrpc_pairs_loss": 0.08277301490306854, + "eval_mrpc_pairs_runtime": 0.2472, + "eval_mrpc_pairs_samples_per_second": 809.045, + "eval_mrpc_pairs_steps_per_second": 20.226, + "step": 12933 + }, + { + "epoch": 2.0326223337515685, + "grad_norm": 8.625104904174805, + "learning_rate": 3.3361195973564865e-05, + "loss": 0.9107, + "step": 12960 + }, + { + "epoch": 2.040150564617315, + "grad_norm": 11.012100219726562, + "learning_rate": 3.320225833821311e-05, + "loss": 0.8694, + "step": 13008 + }, + { + "epoch": 2.0476787954830615, + "grad_norm": 18.39250946044922, + "learning_rate": 3.303638122680463e-05, + "loss": 1.2955, + "step": 13056 + }, + { + "epoch": 2.055207026348808, + "grad_norm": 0.33739256858825684, + "learning_rate": 3.286363794728137e-05, + "loss": 0.7578, + "step": 13104 + }, + { + "epoch": 2.0627352572145545, + "grad_norm": 11.637800216674805, + "learning_rate": 3.268410484202841e-05, + "loss": 0.8352, + "step": 13152 + }, + { + "epoch": 2.0702634880803013, + "grad_norm": 3.6323413848876953, + "learning_rate": 3.249786125413502e-05, + "loss": 1.1223, + "step": 13200 + }, + { + "epoch": 2.0777917189460475, + "grad_norm": 2.320704936981201, + "learning_rate": 3.230498949232972e-05, + "loss": 0.8278, + "step": 13248 + }, + { + "epoch": 2.0853199498117942, + "grad_norm": 19.040969848632812, + "learning_rate": 3.2105574794604686e-05, + "loss": 0.9415, + "step": 13296 + }, + { + "epoch": 2.092848180677541, + "grad_norm": 0.5109882950782776, + "learning_rate": 3.18997052905455e-05, + "loss": 1.017, + "step": 13344 + } + ], + "logging_steps": 48, + "max_steps": 19128, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 1913, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 0.0, + "train_batch_size": 42, + "trial_name": null, + "trial_params": null +}