| { | |
| "best_metric": 0.8734126984126984, | |
| "best_model_checkpoint": "21BAI1229/checkpoint-474", | |
| "epoch": 19.746835443037973, | |
| "eval_steps": 500, | |
| "global_step": 780, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9873417721518988, | |
| "grad_norm": 11.754476547241211, | |
| "learning_rate": 2.5e-05, | |
| "loss": 2.6034, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.9873417721518988, | |
| "eval_accuracy": 0.451984126984127, | |
| "eval_loss": 2.054410696029663, | |
| "eval_runtime": 36.1954, | |
| "eval_samples_per_second": 69.622, | |
| "eval_steps_per_second": 1.105, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 7.275434970855713, | |
| "learning_rate": 4.992877492877493e-05, | |
| "loss": 1.4429, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7849206349206349, | |
| "eval_loss": 0.7735527157783508, | |
| "eval_runtime": 35.4184, | |
| "eval_samples_per_second": 71.149, | |
| "eval_steps_per_second": 1.129, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 2.9873417721518987, | |
| "grad_norm": 7.623991012573242, | |
| "learning_rate": 4.7150997150997157e-05, | |
| "loss": 0.8307, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 2.9873417721518987, | |
| "eval_accuracy": 0.8412698412698413, | |
| "eval_loss": 0.5455929636955261, | |
| "eval_runtime": 35.3707, | |
| "eval_samples_per_second": 71.245, | |
| "eval_steps_per_second": 1.131, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 8.973851203918457, | |
| "learning_rate": 4.4301994301994304e-05, | |
| "loss": 0.6814, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8515873015873016, | |
| "eval_loss": 0.48805657029151917, | |
| "eval_runtime": 35.4085, | |
| "eval_samples_per_second": 71.169, | |
| "eval_steps_per_second": 1.13, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 4.987341772151899, | |
| "grad_norm": 8.185949325561523, | |
| "learning_rate": 4.152421652421652e-05, | |
| "loss": 0.6199, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 4.987341772151899, | |
| "eval_accuracy": 0.8527777777777777, | |
| "eval_loss": 0.46135592460632324, | |
| "eval_runtime": 35.2536, | |
| "eval_samples_per_second": 71.482, | |
| "eval_steps_per_second": 1.135, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 11.136569023132324, | |
| "learning_rate": 3.867521367521368e-05, | |
| "loss": 0.5578, | |
| "step": 237 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8615079365079366, | |
| "eval_loss": 0.44191327691078186, | |
| "eval_runtime": 35.2038, | |
| "eval_samples_per_second": 71.583, | |
| "eval_steps_per_second": 1.136, | |
| "step": 237 | |
| }, | |
| { | |
| "epoch": 6.987341772151899, | |
| "grad_norm": 6.935160160064697, | |
| "learning_rate": 3.58974358974359e-05, | |
| "loss": 0.5198, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 6.987341772151899, | |
| "eval_accuracy": 0.8603174603174604, | |
| "eval_loss": 0.4485108256340027, | |
| "eval_runtime": 35.2921, | |
| "eval_samples_per_second": 71.404, | |
| "eval_steps_per_second": 1.133, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 7.163381576538086, | |
| "learning_rate": 3.304843304843305e-05, | |
| "loss": 0.4811, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8658730158730159, | |
| "eval_loss": 0.4355041980743408, | |
| "eval_runtime": 35.4396, | |
| "eval_samples_per_second": 71.107, | |
| "eval_steps_per_second": 1.129, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 8.987341772151899, | |
| "grad_norm": 7.22255277633667, | |
| "learning_rate": 3.0270655270655275e-05, | |
| "loss": 0.4568, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 8.987341772151899, | |
| "eval_accuracy": 0.8650793650793651, | |
| "eval_loss": 0.4182125926017761, | |
| "eval_runtime": 35.5074, | |
| "eval_samples_per_second": 70.971, | |
| "eval_steps_per_second": 1.127, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 7.7428879737854, | |
| "learning_rate": 2.7421652421652423e-05, | |
| "loss": 0.4268, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8702380952380953, | |
| "eval_loss": 0.4093915522098541, | |
| "eval_runtime": 35.1709, | |
| "eval_samples_per_second": 71.65, | |
| "eval_steps_per_second": 1.137, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 10.987341772151899, | |
| "grad_norm": 8.56812572479248, | |
| "learning_rate": 2.4643874643874645e-05, | |
| "loss": 0.4281, | |
| "step": 434 | |
| }, | |
| { | |
| "epoch": 10.987341772151899, | |
| "eval_accuracy": 0.8706349206349207, | |
| "eval_loss": 0.41577932238578796, | |
| "eval_runtime": 35.2893, | |
| "eval_samples_per_second": 71.41, | |
| "eval_steps_per_second": 1.133, | |
| "step": 434 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 9.711762428283691, | |
| "learning_rate": 2.1794871794871795e-05, | |
| "loss": 0.4143, | |
| "step": 474 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8734126984126984, | |
| "eval_loss": 0.40782999992370605, | |
| "eval_runtime": 35.0211, | |
| "eval_samples_per_second": 71.957, | |
| "eval_steps_per_second": 1.142, | |
| "step": 474 | |
| }, | |
| { | |
| "epoch": 12.987341772151899, | |
| "grad_norm": 7.874723434448242, | |
| "learning_rate": 1.9017094017094017e-05, | |
| "loss": 0.4009, | |
| "step": 513 | |
| }, | |
| { | |
| "epoch": 12.987341772151899, | |
| "eval_accuracy": 0.8714285714285714, | |
| "eval_loss": 0.4066493511199951, | |
| "eval_runtime": 35.2449, | |
| "eval_samples_per_second": 71.5, | |
| "eval_steps_per_second": 1.135, | |
| "step": 513 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 8.416353225708008, | |
| "learning_rate": 1.6168091168091168e-05, | |
| "loss": 0.3642, | |
| "step": 553 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8682539682539683, | |
| "eval_loss": 0.4131360352039337, | |
| "eval_runtime": 35.3914, | |
| "eval_samples_per_second": 71.204, | |
| "eval_steps_per_second": 1.13, | |
| "step": 553 | |
| }, | |
| { | |
| "epoch": 14.987341772151899, | |
| "grad_norm": 8.845190048217773, | |
| "learning_rate": 1.3390313390313392e-05, | |
| "loss": 0.3659, | |
| "step": 592 | |
| }, | |
| { | |
| "epoch": 14.987341772151899, | |
| "eval_accuracy": 0.8726190476190476, | |
| "eval_loss": 0.40469926595687866, | |
| "eval_runtime": 35.2434, | |
| "eval_samples_per_second": 71.503, | |
| "eval_steps_per_second": 1.135, | |
| "step": 592 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 7.056828022003174, | |
| "learning_rate": 1.0541310541310543e-05, | |
| "loss": 0.3487, | |
| "step": 632 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.871031746031746, | |
| "eval_loss": 0.4053677022457123, | |
| "eval_runtime": 35.2106, | |
| "eval_samples_per_second": 71.569, | |
| "eval_steps_per_second": 1.136, | |
| "step": 632 | |
| }, | |
| { | |
| "epoch": 16.9873417721519, | |
| "grad_norm": 7.8862199783325195, | |
| "learning_rate": 7.763532763532765e-06, | |
| "loss": 0.35, | |
| "step": 671 | |
| }, | |
| { | |
| "epoch": 16.9873417721519, | |
| "eval_accuracy": 0.8722222222222222, | |
| "eval_loss": 0.41073036193847656, | |
| "eval_runtime": 35.125, | |
| "eval_samples_per_second": 71.744, | |
| "eval_steps_per_second": 1.139, | |
| "step": 671 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 9.344978332519531, | |
| "learning_rate": 4.914529914529915e-06, | |
| "loss": 0.3291, | |
| "step": 711 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8698412698412699, | |
| "eval_loss": 0.40985915064811707, | |
| "eval_runtime": 35.2658, | |
| "eval_samples_per_second": 71.457, | |
| "eval_steps_per_second": 1.134, | |
| "step": 711 | |
| }, | |
| { | |
| "epoch": 18.9873417721519, | |
| "grad_norm": 6.548698902130127, | |
| "learning_rate": 2.136752136752137e-06, | |
| "loss": 0.338, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 18.9873417721519, | |
| "eval_accuracy": 0.8718253968253968, | |
| "eval_loss": 0.40625905990600586, | |
| "eval_runtime": 35.4023, | |
| "eval_samples_per_second": 71.182, | |
| "eval_steps_per_second": 1.13, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 19.746835443037973, | |
| "grad_norm": 6.30403470993042, | |
| "learning_rate": 0.0, | |
| "loss": 0.3419, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 19.746835443037973, | |
| "eval_accuracy": 0.8702380952380953, | |
| "eval_loss": 0.4066447913646698, | |
| "eval_runtime": 35.3364, | |
| "eval_samples_per_second": 71.315, | |
| "eval_steps_per_second": 1.132, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 19.746835443037973, | |
| "step": 780, | |
| "total_flos": 1.5428282771770638e+19, | |
| "train_loss": 0.6176073722350292, | |
| "train_runtime": 7965.5555, | |
| "train_samples_per_second": 25.309, | |
| "train_steps_per_second": 0.098 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 780, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.5428282771770638e+19, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |