| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 50, | |
| "global_step": 222, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06756756756756757, | |
| "grad_norm": 18.415670170963544, | |
| "learning_rate": 5e-07, | |
| "loss": 1.7484, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.13513513513513514, | |
| "grad_norm": 12.979363840068762, | |
| "learning_rate": 1e-06, | |
| "loss": 1.6, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.20270270270270271, | |
| "grad_norm": 7.702031949468892, | |
| "learning_rate": 9.98628141419305e-07, | |
| "loss": 1.3118, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2702702702702703, | |
| "grad_norm": 3.8773275511437153, | |
| "learning_rate": 9.94520093661082e-07, | |
| "loss": 1.1295, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.33783783783783783, | |
| "grad_norm": 4.237402807418457, | |
| "learning_rate": 9.876983993675989e-07, | |
| "loss": 1.1044, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.40540540540540543, | |
| "grad_norm": 3.642461133839574, | |
| "learning_rate": 9.78200492138261e-07, | |
| "loss": 1.0105, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.47297297297297297, | |
| "grad_norm": 3.608837853456925, | |
| "learning_rate": 9.66078491115194e-07, | |
| "loss": 0.991, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.5405405405405406, | |
| "grad_norm": 3.79449822480748, | |
| "learning_rate": 9.513989149828717e-07, | |
| "loss": 0.9674, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.6081081081081081, | |
| "grad_norm": 3.6384041234003823, | |
| "learning_rate": 9.342423169512071e-07, | |
| "loss": 0.9507, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6756756756756757, | |
| "grad_norm": 3.47968028739859, | |
| "learning_rate": 9.147028427251009e-07, | |
| "loss": 0.9386, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6756756756756757, | |
| "eval_loss": 0.9245508313179016, | |
| "eval_runtime": 36.7866, | |
| "eval_samples_per_second": 57.086, | |
| "eval_steps_per_second": 0.897, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.7432432432432432, | |
| "grad_norm": 3.488579700124589, | |
| "learning_rate": 8.928877138860706e-07, | |
| "loss": 0.946, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.8108108108108109, | |
| "grad_norm": 3.3962328364213157, | |
| "learning_rate": 8.689166395208636e-07, | |
| "loss": 0.9186, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.8783783783783784, | |
| "grad_norm": 3.719808626989404, | |
| "learning_rate": 8.429211593257052e-07, | |
| "loss": 0.9203, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.9459459459459459, | |
| "grad_norm": 3.449706597029694, | |
| "learning_rate": 8.150439217908556e-07, | |
| "loss": 0.9104, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.0135135135135136, | |
| "grad_norm": 3.767188694894111, | |
| "learning_rate": 7.854379014263876e-07, | |
| "loss": 0.8925, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "grad_norm": 3.530860850867838, | |
| "learning_rate": 7.542655593246103e-07, | |
| "loss": 0.8628, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.1486486486486487, | |
| "grad_norm": 3.839676176721922, | |
| "learning_rate": 7.216979516654943e-07, | |
| "loss": 0.8289, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.2162162162162162, | |
| "grad_norm": 3.629466135685692, | |
| "learning_rate": 6.87913791057119e-07, | |
| "loss": 0.8346, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.2837837837837838, | |
| "grad_norm": 3.758252497307305, | |
| "learning_rate": 6.530984658619733e-07, | |
| "loss": 0.8442, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.3513513513513513, | |
| "grad_norm": 3.6510886764874098, | |
| "learning_rate": 6.174430228904919e-07, | |
| "loss": 0.8263, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.3513513513513513, | |
| "eval_loss": 0.8765044808387756, | |
| "eval_runtime": 36.4454, | |
| "eval_samples_per_second": 57.62, | |
| "eval_steps_per_second": 0.905, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.4189189189189189, | |
| "grad_norm": 3.5720282153519287, | |
| "learning_rate": 5.8114311904423e-07, | |
| "loss": 0.8266, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.4864864864864864, | |
| "grad_norm": 3.362670058366568, | |
| "learning_rate": 5.443979476614674e-07, | |
| "loss": 0.8094, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.554054054054054, | |
| "grad_norm": 3.725544496637925, | |
| "learning_rate": 5.074091454568463e-07, | |
| "loss": 0.8107, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.6216216216216215, | |
| "grad_norm": 3.399706437395112, | |
| "learning_rate": 4.703796860531429e-07, | |
| "loss": 0.8112, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.689189189189189, | |
| "grad_norm": 3.4945021536853718, | |
| "learning_rate": 4.3351276617684285e-07, | |
| "loss": 0.8197, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.7567567567567568, | |
| "grad_norm": 3.423613980661302, | |
| "learning_rate": 3.970106906294509e-07, | |
| "loss": 0.8158, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.8243243243243243, | |
| "grad_norm": 3.4537863970020117, | |
| "learning_rate": 3.610737621531781e-07, | |
| "loss": 0.8297, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.8918918918918919, | |
| "grad_norm": 3.4702507620367875, | |
| "learning_rate": 3.2589918228280066e-07, | |
| "loss": 0.8031, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.9594594594594594, | |
| "grad_norm": 3.3271823791904884, | |
| "learning_rate": 2.916799692151884e-07, | |
| "loss": 0.8016, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 2.027027027027027, | |
| "grad_norm": 3.9798718651465848, | |
| "learning_rate": 2.5860389863462763e-07, | |
| "loss": 0.7963, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.027027027027027, | |
| "eval_loss": 0.8589270710945129, | |
| "eval_runtime": 36.48, | |
| "eval_samples_per_second": 57.566, | |
| "eval_steps_per_second": 0.905, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.0945945945945947, | |
| "grad_norm": 3.542726689032495, | |
| "learning_rate": 2.2685247330608414e-07, | |
| "loss": 0.7712, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 2.1621621621621623, | |
| "grad_norm": 3.4799701202917244, | |
| "learning_rate": 1.9659992709070344e-07, | |
| "loss": 0.7662, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.22972972972973, | |
| "grad_norm": 3.594791003920676, | |
| "learning_rate": 1.6801226884893893e-07, | |
| "loss": 0.7538, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 2.2972972972972974, | |
| "grad_norm": 3.5480210206929637, | |
| "learning_rate": 1.412463714778343e-07, | |
| "loss": 0.7731, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.364864864864865, | |
| "grad_norm": 3.5836803836525606, | |
| "learning_rate": 1.1644911108130434e-07, | |
| "loss": 0.7792, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.4324324324324325, | |
| "grad_norm": 3.6501667972586693, | |
| "learning_rate": 9.375656099715934e-08, | |
| "loss": 0.757, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 3.686290348200621, | |
| "learning_rate": 7.329324510360269e-08, | |
| "loss": 0.7426, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 2.5675675675675675, | |
| "grad_norm": 3.574089956841002, | |
| "learning_rate": 5.517145450262639e-08, | |
| "loss": 0.754, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.635135135135135, | |
| "grad_norm": 3.6761723022132617, | |
| "learning_rate": 3.9490631329964554e-08, | |
| "loss": 0.7604, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 2.7027027027027026, | |
| "grad_norm": 3.5287319124762004, | |
| "learning_rate": 2.63368230729043e-08, | |
| "loss": 0.7574, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.7027027027027026, | |
| "eval_loss": 0.8600885272026062, | |
| "eval_runtime": 36.5528, | |
| "eval_samples_per_second": 57.451, | |
| "eval_steps_per_second": 0.903, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.77027027027027, | |
| "grad_norm": 3.4745956970229845, | |
| "learning_rate": 1.5782210390350713e-08, | |
| "loss": 0.7485, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 2.8378378378378377, | |
| "grad_norm": 3.5410239237454446, | |
| "learning_rate": 7.884711026201584e-09, | |
| "loss": 0.7526, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.9054054054054053, | |
| "grad_norm": 3.717018997325894, | |
| "learning_rate": 2.687661989531964e-09, | |
| "loss": 0.7473, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 2.972972972972973, | |
| "grad_norm": 3.6243019256758044, | |
| "learning_rate": 2.1958174560282594e-10, | |
| "loss": 0.7482, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 222, | |
| "total_flos": 1308778674782208.0, | |
| "train_loss": 0.8914520751248609, | |
| "train_runtime": 3092.5546, | |
| "train_samples_per_second": 18.334, | |
| "train_steps_per_second": 0.072 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 222, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 10000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1308778674782208.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |