| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.974948758824869, | |
| "global_step": 10950, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.553734061930783e-05, | |
| "loss": 4.5455, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "eval_loss": 0.06012005731463432, | |
| "eval_runtime": 125.7302, | |
| "eval_samples_per_second": 19.86, | |
| "eval_steps_per_second": 1.249, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 9.107468123861566e-05, | |
| "loss": 0.0471, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_loss": 0.05808735638856888, | |
| "eval_runtime": 125.3179, | |
| "eval_samples_per_second": 19.925, | |
| "eval_steps_per_second": 1.253, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 9.807120237981e-05, | |
| "loss": 0.0443, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_loss": 0.05481741577386856, | |
| "eval_runtime": 125.3239, | |
| "eval_samples_per_second": 19.924, | |
| "eval_steps_per_second": 1.253, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 9.567220036464831e-05, | |
| "loss": 0.0428, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "eval_loss": 0.05039665102958679, | |
| "eval_runtime": 125.3169, | |
| "eval_samples_per_second": 19.925, | |
| "eval_steps_per_second": 1.253, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 9.327319834948663e-05, | |
| "loss": 0.0329, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_loss": 0.05058171600103378, | |
| "eval_runtime": 125.3095, | |
| "eval_samples_per_second": 19.927, | |
| "eval_steps_per_second": 1.253, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 9.087419633432492e-05, | |
| "loss": 0.0299, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "eval_loss": 0.04842585325241089, | |
| "eval_runtime": 125.3053, | |
| "eval_samples_per_second": 19.927, | |
| "eval_steps_per_second": 1.253, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 8.847519431916324e-05, | |
| "loss": 0.0295, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "eval_loss": 0.049905285239219666, | |
| "eval_runtime": 125.3158, | |
| "eval_samples_per_second": 19.926, | |
| "eval_steps_per_second": 1.253, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 8.607619230400153e-05, | |
| "loss": 0.0304, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_loss": 0.047520652413368225, | |
| "eval_runtime": 125.3111, | |
| "eval_samples_per_second": 19.926, | |
| "eval_steps_per_second": 1.253, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 8.367719028883985e-05, | |
| "loss": 0.0277, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "eval_loss": 0.04981054365634918, | |
| "eval_runtime": 125.3207, | |
| "eval_samples_per_second": 19.925, | |
| "eval_steps_per_second": 1.253, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 8.127818827367816e-05, | |
| "loss": 0.0186, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "eval_loss": 0.04637761414051056, | |
| "eval_runtime": 125.3226, | |
| "eval_samples_per_second": 19.925, | |
| "eval_steps_per_second": 1.253, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 7.887918625851645e-05, | |
| "loss": 0.0199, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "eval_loss": 0.04944201186299324, | |
| "eval_runtime": 125.3132, | |
| "eval_samples_per_second": 19.926, | |
| "eval_steps_per_second": 1.253, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 7.648018424335477e-05, | |
| "loss": 0.0205, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "eval_loss": 0.045388150960206985, | |
| "eval_runtime": 125.3269, | |
| "eval_samples_per_second": 19.924, | |
| "eval_steps_per_second": 1.253, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 7.408118222819308e-05, | |
| "loss": 0.0202, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_loss": 0.04644118994474411, | |
| "eval_runtime": 125.317, | |
| "eval_samples_per_second": 19.925, | |
| "eval_steps_per_second": 1.253, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 7.168218021303138e-05, | |
| "loss": 0.015, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "eval_loss": 0.0492834635078907, | |
| "eval_runtime": 125.3122, | |
| "eval_samples_per_second": 19.926, | |
| "eval_steps_per_second": 1.253, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 6.928317819786969e-05, | |
| "loss": 0.0136, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "eval_loss": 0.050950221717357635, | |
| "eval_runtime": 125.3101, | |
| "eval_samples_per_second": 19.927, | |
| "eval_steps_per_second": 1.253, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 6.6884176182708e-05, | |
| "loss": 0.0139, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "eval_loss": 0.0521300733089447, | |
| "eval_runtime": 125.3091, | |
| "eval_samples_per_second": 19.927, | |
| "eval_steps_per_second": 1.253, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 6.44851741675463e-05, | |
| "loss": 0.0149, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "eval_loss": 0.049005962908267975, | |
| "eval_runtime": 125.2924, | |
| "eval_samples_per_second": 19.929, | |
| "eval_steps_per_second": 1.253, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 6.208617215238462e-05, | |
| "loss": 0.012, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "eval_loss": 0.05201614275574684, | |
| "eval_runtime": 125.3236, | |
| "eval_samples_per_second": 19.924, | |
| "eval_steps_per_second": 1.253, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "learning_rate": 5.968717013722291e-05, | |
| "loss": 0.0094, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "eval_loss": 0.052882954478263855, | |
| "eval_runtime": 125.3182, | |
| "eval_samples_per_second": 19.925, | |
| "eval_steps_per_second": 1.253, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 5.7288168122061226e-05, | |
| "loss": 0.0104, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "eval_loss": 0.05443257838487625, | |
| "eval_runtime": 125.308, | |
| "eval_samples_per_second": 19.927, | |
| "eval_steps_per_second": 1.253, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 5.488916610689954e-05, | |
| "loss": 0.0095, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "eval_loss": 0.0512896366417408, | |
| "eval_runtime": 125.3186, | |
| "eval_samples_per_second": 19.925, | |
| "eval_steps_per_second": 1.253, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 5.249016409173784e-05, | |
| "loss": 0.0099, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "eval_loss": 0.05250149220228195, | |
| "eval_runtime": 125.3202, | |
| "eval_samples_per_second": 19.925, | |
| "eval_steps_per_second": 1.253, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 5.009116207657615e-05, | |
| "loss": 0.0067, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "eval_loss": 0.05294517055153847, | |
| "eval_runtime": 125.3036, | |
| "eval_samples_per_second": 19.928, | |
| "eval_steps_per_second": 1.253, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 5.47, | |
| "learning_rate": 4.769216006141446e-05, | |
| "loss": 0.0064, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 5.47, | |
| "eval_loss": 0.05718787759542465, | |
| "eval_runtime": 125.3174, | |
| "eval_samples_per_second": 19.925, | |
| "eval_steps_per_second": 1.253, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 4.5293158046252756e-05, | |
| "loss": 0.0062, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "eval_loss": 0.0588238462805748, | |
| "eval_runtime": 125.32, | |
| "eval_samples_per_second": 19.925, | |
| "eval_steps_per_second": 1.253, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 4.289415603109107e-05, | |
| "loss": 0.0066, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "eval_loss": 0.055590804666280746, | |
| "eval_runtime": 125.32, | |
| "eval_samples_per_second": 19.925, | |
| "eval_steps_per_second": 1.253, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 4.0495154015929375e-05, | |
| "loss": 0.0049, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "eval_loss": 0.060405001044273376, | |
| "eval_runtime": 125.3175, | |
| "eval_samples_per_second": 19.925, | |
| "eval_steps_per_second": 1.253, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 6.38, | |
| "learning_rate": 3.809615200076768e-05, | |
| "loss": 0.0044, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 6.38, | |
| "eval_loss": 0.0592646524310112, | |
| "eval_runtime": 125.3136, | |
| "eval_samples_per_second": 19.926, | |
| "eval_steps_per_second": 1.253, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 6.6, | |
| "learning_rate": 3.569714998560599e-05, | |
| "loss": 0.0042, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 6.6, | |
| "eval_loss": 0.059081513434648514, | |
| "eval_runtime": 125.2994, | |
| "eval_samples_per_second": 19.928, | |
| "eval_steps_per_second": 1.253, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 3.32981479704443e-05, | |
| "loss": 0.0048, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "eval_loss": 0.06123210862278938, | |
| "eval_runtime": 125.3056, | |
| "eval_samples_per_second": 19.927, | |
| "eval_steps_per_second": 1.253, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 3.0899145955282606e-05, | |
| "loss": 0.004, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "eval_loss": 0.060906291007995605, | |
| "eval_runtime": 125.3113, | |
| "eval_samples_per_second": 19.926, | |
| "eval_steps_per_second": 1.253, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 7.29, | |
| "learning_rate": 2.850014394012091e-05, | |
| "loss": 0.003, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 7.29, | |
| "eval_loss": 0.06742047518491745, | |
| "eval_runtime": 125.3117, | |
| "eval_samples_per_second": 19.926, | |
| "eval_steps_per_second": 1.253, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "learning_rate": 2.6101141924959215e-05, | |
| "loss": 0.003, | |
| "step": 8250 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "eval_loss": 0.0640687569975853, | |
| "eval_runtime": 125.31, | |
| "eval_samples_per_second": 19.927, | |
| "eval_steps_per_second": 1.253, | |
| "step": 8250 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 2.3702139909797524e-05, | |
| "loss": 0.0027, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "eval_loss": 0.06774434447288513, | |
| "eval_runtime": 125.316, | |
| "eval_samples_per_second": 19.926, | |
| "eval_steps_per_second": 1.253, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "learning_rate": 2.1303137894635834e-05, | |
| "loss": 0.0028, | |
| "step": 8750 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "eval_loss": 0.06737840920686722, | |
| "eval_runtime": 125.323, | |
| "eval_samples_per_second": 19.925, | |
| "eval_steps_per_second": 1.253, | |
| "step": 8750 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "learning_rate": 1.890413587947414e-05, | |
| "loss": 0.0021, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "eval_loss": 0.06941425800323486, | |
| "eval_runtime": 125.3147, | |
| "eval_samples_per_second": 19.926, | |
| "eval_steps_per_second": 1.253, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 8.43, | |
| "learning_rate": 1.6505133864312446e-05, | |
| "loss": 0.0018, | |
| "step": 9250 | |
| }, | |
| { | |
| "epoch": 8.43, | |
| "eval_loss": 0.07149343937635422, | |
| "eval_runtime": 125.3155, | |
| "eval_samples_per_second": 19.926, | |
| "eval_steps_per_second": 1.253, | |
| "step": 9250 | |
| }, | |
| { | |
| "epoch": 8.65, | |
| "learning_rate": 1.4106131849150753e-05, | |
| "loss": 0.0021, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 8.65, | |
| "eval_loss": 0.06807977706193924, | |
| "eval_runtime": 125.3164, | |
| "eval_samples_per_second": 19.926, | |
| "eval_steps_per_second": 1.253, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 8.88, | |
| "learning_rate": 1.1707129833989061e-05, | |
| "loss": 0.0017, | |
| "step": 9750 | |
| }, | |
| { | |
| "epoch": 8.88, | |
| "eval_loss": 0.07044515013694763, | |
| "eval_runtime": 125.3307, | |
| "eval_samples_per_second": 19.923, | |
| "eval_steps_per_second": 1.253, | |
| "step": 9750 | |
| }, | |
| { | |
| "epoch": 9.11, | |
| "learning_rate": 9.308127818827369e-06, | |
| "loss": 0.0014, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 9.11, | |
| "eval_loss": 0.07252407819032669, | |
| "eval_runtime": 125.329, | |
| "eval_samples_per_second": 19.924, | |
| "eval_steps_per_second": 1.253, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 9.34, | |
| "learning_rate": 6.909125803665675e-06, | |
| "loss": 0.0012, | |
| "step": 10250 | |
| }, | |
| { | |
| "epoch": 9.34, | |
| "eval_loss": 0.07298342883586884, | |
| "eval_runtime": 125.3163, | |
| "eval_samples_per_second": 19.926, | |
| "eval_steps_per_second": 1.253, | |
| "step": 10250 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "learning_rate": 4.510123788503983e-06, | |
| "loss": 0.0012, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "eval_loss": 0.07320970296859741, | |
| "eval_runtime": 125.318, | |
| "eval_samples_per_second": 19.925, | |
| "eval_steps_per_second": 1.253, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "learning_rate": 2.11112177334229e-06, | |
| "loss": 0.0013, | |
| "step": 10750 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "eval_loss": 0.07287949323654175, | |
| "eval_runtime": 125.3257, | |
| "eval_samples_per_second": 19.924, | |
| "eval_steps_per_second": 1.253, | |
| "step": 10750 | |
| } | |
| ], | |
| "max_steps": 10970, | |
| "num_train_epochs": 10, | |
| "total_flos": 2.0454829414889472e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |