| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5714285714285714, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005714285714285714, | |
| "eval_loss": 1.3168718814849854, | |
| "eval_runtime": 40.692, | |
| "eval_samples_per_second": 7.25, | |
| "eval_steps_per_second": 0.909, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.017142857142857144, | |
| "grad_norm": 0.6075395941734314, | |
| "learning_rate": 3e-05, | |
| "loss": 1.232, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.03428571428571429, | |
| "grad_norm": 0.6195719838142395, | |
| "learning_rate": 6e-05, | |
| "loss": 1.2843, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.05142857142857143, | |
| "grad_norm": 0.5386181473731995, | |
| "learning_rate": 9e-05, | |
| "loss": 1.227, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.05142857142857143, | |
| "eval_loss": 1.2125165462493896, | |
| "eval_runtime": 41.5705, | |
| "eval_samples_per_second": 7.096, | |
| "eval_steps_per_second": 0.89, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.06857142857142857, | |
| "grad_norm": 0.5988479852676392, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.1529, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.08571428571428572, | |
| "grad_norm": 0.4145963191986084, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 1.1147, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.10285714285714286, | |
| "grad_norm": 0.3078181743621826, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 1.0461, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.10285714285714286, | |
| "eval_loss": 1.1064488887786865, | |
| "eval_runtime": 41.5719, | |
| "eval_samples_per_second": 7.096, | |
| "eval_steps_per_second": 0.89, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.403085857629776, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.9892, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.13714285714285715, | |
| "grad_norm": 0.3588714301586151, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 1.0679, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.15428571428571428, | |
| "grad_norm": 0.349284291267395, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.9949, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.15428571428571428, | |
| "eval_loss": 1.0255528688430786, | |
| "eval_runtime": 41.5637, | |
| "eval_samples_per_second": 7.098, | |
| "eval_steps_per_second": 0.89, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.17142857142857143, | |
| "grad_norm": 0.33968040347099304, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.9591, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.18857142857142858, | |
| "grad_norm": 0.3603692948818207, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.9444, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.2057142857142857, | |
| "grad_norm": 0.3915916979312897, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.9726, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.2057142857142857, | |
| "eval_loss": 0.9834849238395691, | |
| "eval_runtime": 41.554, | |
| "eval_samples_per_second": 7.099, | |
| "eval_steps_per_second": 0.89, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.22285714285714286, | |
| "grad_norm": 0.41616091132164, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 1.0561, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.29904600977897644, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.8935, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.2571428571428571, | |
| "grad_norm": 0.348360151052475, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.9165, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2571428571428571, | |
| "eval_loss": 0.9552392959594727, | |
| "eval_runtime": 41.563, | |
| "eval_samples_per_second": 7.098, | |
| "eval_steps_per_second": 0.89, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2742857142857143, | |
| "grad_norm": 0.39516714215278625, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.9499, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.2914285714285714, | |
| "grad_norm": 0.3064298927783966, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.9638, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.30857142857142855, | |
| "grad_norm": 0.398806631565094, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.9591, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.30857142857142855, | |
| "eval_loss": 0.9361926317214966, | |
| "eval_runtime": 41.526, | |
| "eval_samples_per_second": 7.104, | |
| "eval_steps_per_second": 0.891, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.32571428571428573, | |
| "grad_norm": 0.38757139444351196, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.8849, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.34285714285714286, | |
| "grad_norm": 0.38670626282691956, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 1.0235, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.33959320187568665, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.9051, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 0.9213638305664062, | |
| "eval_runtime": 41.504, | |
| "eval_samples_per_second": 7.108, | |
| "eval_steps_per_second": 0.891, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.37714285714285717, | |
| "grad_norm": 0.36085259914398193, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.9088, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.3942857142857143, | |
| "grad_norm": 0.4006972014904022, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.9164, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.4114285714285714, | |
| "grad_norm": 0.3791073262691498, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.8795, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.4114285714285714, | |
| "eval_loss": 0.9112246632575989, | |
| "eval_runtime": 41.5061, | |
| "eval_samples_per_second": 7.107, | |
| "eval_steps_per_second": 0.891, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.42857142857142855, | |
| "grad_norm": 0.3648053705692291, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.8716, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.44571428571428573, | |
| "grad_norm": 0.4187893569469452, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.8903, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.46285714285714286, | |
| "grad_norm": 0.3176676630973816, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 0.963, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.46285714285714286, | |
| "eval_loss": 0.9049162268638611, | |
| "eval_runtime": 41.4996, | |
| "eval_samples_per_second": 7.108, | |
| "eval_steps_per_second": 0.892, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.305603563785553, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.8787, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.49714285714285716, | |
| "grad_norm": 0.38687825202941895, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.9204, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.5142857142857142, | |
| "grad_norm": 0.3953753709793091, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 1.0274, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5142857142857142, | |
| "eval_loss": 0.9023739695549011, | |
| "eval_runtime": 41.5212, | |
| "eval_samples_per_second": 7.105, | |
| "eval_steps_per_second": 0.891, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5314285714285715, | |
| "grad_norm": 0.37336939573287964, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.9771, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.5485714285714286, | |
| "grad_norm": 0.5363125801086426, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.8811, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.5657142857142857, | |
| "grad_norm": 0.39761680364608765, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 0.8971, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.5657142857142857, | |
| "eval_loss": 0.9018054008483887, | |
| "eval_runtime": 41.5636, | |
| "eval_samples_per_second": 7.098, | |
| "eval_steps_per_second": 0.89, | |
| "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": 2.9267453336459674e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |