| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.42598509052183176, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004259850905218318, | |
| "eval_loss": 1.2758299112319946, | |
| "eval_runtime": 10.863, | |
| "eval_samples_per_second": 36.362, | |
| "eval_steps_per_second": 4.603, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.012779552715654952, | |
| "grad_norm": 0.8314210176467896, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2924, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.025559105431309903, | |
| "grad_norm": 0.8669793009757996, | |
| "learning_rate": 6e-05, | |
| "loss": 1.2805, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.038338658146964855, | |
| "grad_norm": 1.0953078269958496, | |
| "learning_rate": 9e-05, | |
| "loss": 1.175, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.038338658146964855, | |
| "eval_loss": 1.003463864326477, | |
| "eval_runtime": 10.8921, | |
| "eval_samples_per_second": 36.265, | |
| "eval_steps_per_second": 4.59, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.051118210862619806, | |
| "grad_norm": 1.2419706583023071, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.017, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.06389776357827476, | |
| "grad_norm": 0.8883413672447205, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.9621, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.07667731629392971, | |
| "grad_norm": 1.3165100812911987, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.7864, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.07667731629392971, | |
| "eval_loss": 0.7199711799621582, | |
| "eval_runtime": 10.8865, | |
| "eval_samples_per_second": 36.284, | |
| "eval_steps_per_second": 4.593, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.08945686900958466, | |
| "grad_norm": 0.6705540418624878, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.7165, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.10223642172523961, | |
| "grad_norm": 0.9872334003448486, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.6824, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.11501597444089456, | |
| "grad_norm": 0.7229445576667786, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.7033, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.11501597444089456, | |
| "eval_loss": 0.6306272745132446, | |
| "eval_runtime": 10.9156, | |
| "eval_samples_per_second": 36.187, | |
| "eval_steps_per_second": 4.581, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.12779552715654952, | |
| "grad_norm": 0.7949721217155457, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.6592, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.14057507987220447, | |
| "grad_norm": 0.5070735812187195, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.6234, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.15335463258785942, | |
| "grad_norm": 0.6172641515731812, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.6746, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.15335463258785942, | |
| "eval_loss": 0.6016714572906494, | |
| "eval_runtime": 10.9227, | |
| "eval_samples_per_second": 36.163, | |
| "eval_steps_per_second": 4.578, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.16613418530351437, | |
| "grad_norm": 0.7518579959869385, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.6258, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.17891373801916932, | |
| "grad_norm": 0.5821406841278076, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.587, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.19169329073482427, | |
| "grad_norm": 0.8557337522506714, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.6309, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.19169329073482427, | |
| "eval_loss": 0.5868030190467834, | |
| "eval_runtime": 10.9092, | |
| "eval_samples_per_second": 36.208, | |
| "eval_steps_per_second": 4.583, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.20447284345047922, | |
| "grad_norm": 0.6080703735351562, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.5596, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.21725239616613418, | |
| "grad_norm": 0.6161754727363586, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.6585, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.23003194888178913, | |
| "grad_norm": 0.682461142539978, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.5783, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.23003194888178913, | |
| "eval_loss": 0.5760997533798218, | |
| "eval_runtime": 10.904, | |
| "eval_samples_per_second": 36.225, | |
| "eval_steps_per_second": 4.585, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.24281150159744408, | |
| "grad_norm": 0.7546740174293518, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.6759, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.25559105431309903, | |
| "grad_norm": 0.7903650403022766, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.6119, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.268370607028754, | |
| "grad_norm": 0.5551833510398865, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.5778, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.268370607028754, | |
| "eval_loss": 0.5682376027107239, | |
| "eval_runtime": 10.9169, | |
| "eval_samples_per_second": 36.182, | |
| "eval_steps_per_second": 4.58, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.28115015974440893, | |
| "grad_norm": 0.6363338828086853, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.5992, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.2939297124600639, | |
| "grad_norm": 0.60361248254776, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.5925, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.30670926517571884, | |
| "grad_norm": 0.7037025690078735, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.6792, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.30670926517571884, | |
| "eval_loss": 0.5637584328651428, | |
| "eval_runtime": 10.9073, | |
| "eval_samples_per_second": 36.214, | |
| "eval_steps_per_second": 4.584, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.3194888178913738, | |
| "grad_norm": 0.7148558497428894, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.6001, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.33226837060702874, | |
| "grad_norm": 0.5377858877182007, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.5808, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.3450479233226837, | |
| "grad_norm": 0.4729587733745575, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 0.517, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.3450479233226837, | |
| "eval_loss": 0.5606918931007385, | |
| "eval_runtime": 10.9225, | |
| "eval_samples_per_second": 36.164, | |
| "eval_steps_per_second": 4.578, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.35782747603833864, | |
| "grad_norm": 0.5885350108146667, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.5907, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.3706070287539936, | |
| "grad_norm": 0.6870759725570679, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.5172, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.38338658146964855, | |
| "grad_norm": 0.5195698142051697, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.5392, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.38338658146964855, | |
| "eval_loss": 0.5600429773330688, | |
| "eval_runtime": 10.9113, | |
| "eval_samples_per_second": 36.201, | |
| "eval_steps_per_second": 4.582, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3961661341853035, | |
| "grad_norm": 0.5903887152671814, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.6567, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.40894568690095845, | |
| "grad_norm": 0.5635388493537903, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.587, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.4217252396166134, | |
| "grad_norm": 0.5114682912826538, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 0.5674, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.4217252396166134, | |
| "eval_loss": 0.5595129132270813, | |
| "eval_runtime": 10.9155, | |
| "eval_samples_per_second": 36.187, | |
| "eval_steps_per_second": 4.581, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.61253767364608e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |