| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.41841004184100417, | |
| "eval_steps": 300, | |
| "global_step": 1200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.017433751743375175, | |
| "grad_norm": 11.72121810913086, | |
| "learning_rate": 9.966527196652718e-07, | |
| "loss": 10.429, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.03486750348675035, | |
| "grad_norm": 7.926635265350342, | |
| "learning_rate": 9.933054393305438e-07, | |
| "loss": 9.5621, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.05230125523012552, | |
| "grad_norm": 14.6965970993042, | |
| "learning_rate": 9.898884239888422e-07, | |
| "loss": 8.9643, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.0697350069735007, | |
| "grad_norm": 19.375049591064453, | |
| "learning_rate": 9.864714086471408e-07, | |
| "loss": 8.4059, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08716875871687588, | |
| "grad_norm": 18.03301429748535, | |
| "learning_rate": 9.833333333333332e-07, | |
| "loss": 7.612, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.10460251046025104, | |
| "grad_norm": 10.96057415008545, | |
| "learning_rate": 9.798465829846583e-07, | |
| "loss": 6.8137, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.10460251046025104, | |
| "eval_gen_len": 10.4739, | |
| "eval_loss": 6.225385665893555, | |
| "eval_rouge1": 6.5635, | |
| "eval_rouge2": 1.1205, | |
| "eval_rougeL": 5.1919, | |
| "eval_rougeLsum": 5.8048, | |
| "eval_runtime": 493.488, | |
| "eval_samples_per_second": 11.036, | |
| "eval_steps_per_second": 0.087, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.12203626220362622, | |
| "grad_norm": 13.111268997192383, | |
| "learning_rate": 9.763598326359831e-07, | |
| "loss": 6.0284, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.1394700139470014, | |
| "grad_norm": 9.499017715454102, | |
| "learning_rate": 9.730822873082288e-07, | |
| "loss": 5.3582, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.15690376569037656, | |
| "grad_norm": 5.717628002166748, | |
| "learning_rate": 9.696652719665272e-07, | |
| "loss": 4.8628, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.17433751743375175, | |
| "grad_norm": 3.644038438796997, | |
| "learning_rate": 9.662482566248256e-07, | |
| "loss": 4.6036, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.19177126917712692, | |
| "grad_norm": 4.016868591308594, | |
| "learning_rate": 9.62831241283124e-07, | |
| "loss": 4.4095, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.20920502092050208, | |
| "grad_norm": 1.7940410375595093, | |
| "learning_rate": 9.59344490934449e-07, | |
| "loss": 4.3409, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.20920502092050208, | |
| "eval_gen_len": 21.2969, | |
| "eval_loss": 4.094496250152588, | |
| "eval_rouge1": 7.0701, | |
| "eval_rouge2": 1.1979, | |
| "eval_rougeL": 5.5075, | |
| "eval_rougeLsum": 6.2106, | |
| "eval_runtime": 464.764, | |
| "eval_samples_per_second": 11.718, | |
| "eval_steps_per_second": 0.093, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.22663877266387727, | |
| "grad_norm": 1.70875084400177, | |
| "learning_rate": 9.558577405857741e-07, | |
| "loss": 4.3321, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.24407252440725244, | |
| "grad_norm": 1.239980936050415, | |
| "learning_rate": 9.52510460251046e-07, | |
| "loss": 4.2697, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.2615062761506276, | |
| "grad_norm": 1.91974675655365, | |
| "learning_rate": 9.49163179916318e-07, | |
| "loss": 4.2407, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.2789400278940028, | |
| "grad_norm": 2.3624725341796875, | |
| "learning_rate": 9.4581589958159e-07, | |
| "loss": 4.2552, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.296373779637378, | |
| "grad_norm": 3.15625, | |
| "learning_rate": 9.424686192468618e-07, | |
| "loss": 4.2662, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.3138075313807531, | |
| "grad_norm": 3.3393449783325195, | |
| "learning_rate": 9.390516039051604e-07, | |
| "loss": 4.2447, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.3138075313807531, | |
| "eval_gen_len": 40.1348, | |
| "eval_loss": 4.038403034210205, | |
| "eval_rouge1": 8.106, | |
| "eval_rouge2": 1.3148, | |
| "eval_rougeL": 6.3905, | |
| "eval_rougeLsum": 7.0688, | |
| "eval_runtime": 806.4671, | |
| "eval_samples_per_second": 6.753, | |
| "eval_steps_per_second": 0.053, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.3312412831241283, | |
| "grad_norm": 4.088630199432373, | |
| "learning_rate": 9.357043235704322e-07, | |
| "loss": 4.2266, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.3486750348675035, | |
| "grad_norm": 0.0, | |
| "learning_rate": 9.324965132496513e-07, | |
| "loss": 4.2561, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.36610878661087864, | |
| "grad_norm": 0.0, | |
| "learning_rate": 9.290794979079497e-07, | |
| "loss": 4.2302, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.38354253835425384, | |
| "grad_norm": 0.0, | |
| "learning_rate": 9.256624825662482e-07, | |
| "loss": 4.2562, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.40097629009762903, | |
| "grad_norm": 0.0, | |
| "learning_rate": 9.223152022315202e-07, | |
| "loss": 4.2719, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.41841004184100417, | |
| "grad_norm": 0.0, | |
| "learning_rate": 9.188981868898187e-07, | |
| "loss": 4.2771, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.41841004184100417, | |
| "eval_gen_len": 40.3153, | |
| "eval_loss": 4.038360118865967, | |
| "eval_rouge1": 8.0991, | |
| "eval_rouge2": 1.3122, | |
| "eval_rougeL": 6.386, | |
| "eval_rougeLsum": 7.0615, | |
| "eval_runtime": 808.1023, | |
| "eval_samples_per_second": 6.739, | |
| "eval_steps_per_second": 0.053, | |
| "step": 1200 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 14340, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 300, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.3383982997504e+16, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |