| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 2288, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08741258741258741, | |
| "grad_norm": 1.2647322416305542, | |
| "learning_rate": 3.4615384615384617e-06, | |
| "loss": 0.6795, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.17482517482517482, | |
| "grad_norm": 2.8214495182037354, | |
| "learning_rate": 6.958041958041959e-06, | |
| "loss": 0.4519, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.26223776223776224, | |
| "grad_norm": 2.6608622074127197, | |
| "learning_rate": 1.0454545454545455e-05, | |
| "loss": 0.3745, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.34965034965034963, | |
| "grad_norm": 2.5102548599243164, | |
| "learning_rate": 1.3951048951048953e-05, | |
| "loss": 0.3628, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.4370629370629371, | |
| "grad_norm": 3.80678129196167, | |
| "learning_rate": 1.744755244755245e-05, | |
| "loss": 0.3649, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5244755244755245, | |
| "grad_norm": 4.311079025268555, | |
| "learning_rate": 1.9998642592088543e-05, | |
| "loss": 0.3665, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.6118881118881119, | |
| "grad_norm": 1.8182202577590942, | |
| "learning_rate": 1.9969981933354087e-05, | |
| "loss": 0.3577, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.6993006993006993, | |
| "grad_norm": 1.7828468084335327, | |
| "learning_rate": 1.9904203472172682e-05, | |
| "loss": 0.363, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.7867132867132867, | |
| "grad_norm": 1.9733550548553467, | |
| "learning_rate": 1.980155209885043e-05, | |
| "loss": 0.356, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.8741258741258742, | |
| "grad_norm": 1.8030153512954712, | |
| "learning_rate": 1.9662409979916944e-05, | |
| "loss": 0.3557, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.9615384615384616, | |
| "grad_norm": 1.6669570207595825, | |
| "learning_rate": 1.9487295135336273e-05, | |
| "loss": 0.3478, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_f1": 0.8321883058258712, | |
| "eval_loss": 0.3490952253341675, | |
| "eval_runtime": 34.3414, | |
| "eval_samples_per_second": 1065.098, | |
| "eval_steps_per_second": 8.328, | |
| "step": 1144 | |
| }, | |
| { | |
| "epoch": 1.048951048951049, | |
| "grad_norm": 1.8223475217819214, | |
| "learning_rate": 1.927685950994143e-05, | |
| "loss": 0.3356, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.1363636363636362, | |
| "grad_norm": 2.0821902751922607, | |
| "learning_rate": 1.9031886546272573e-05, | |
| "loss": 0.3347, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.2237762237762237, | |
| "grad_norm": 1.8656470775604248, | |
| "learning_rate": 1.8753288267855023e-05, | |
| "loss": 0.3273, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.3111888111888113, | |
| "grad_norm": 3.0232980251312256, | |
| "learning_rate": 1.844210188377597e-05, | |
| "loss": 0.3288, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.3986013986013985, | |
| "grad_norm": 3.8725271224975586, | |
| "learning_rate": 1.809948592720084e-05, | |
| "loss": 0.3268, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.486013986013986, | |
| "grad_norm": 3.690535306930542, | |
| "learning_rate": 1.7726715942205463e-05, | |
| "loss": 0.3319, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.5734265734265733, | |
| "grad_norm": 1.3706207275390625, | |
| "learning_rate": 1.732517973498184e-05, | |
| "loss": 0.3312, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.6608391608391608, | |
| "grad_norm": 1.8885810375213623, | |
| "learning_rate": 1.6896372207096963e-05, | |
| "loss": 0.3244, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.7482517482517483, | |
| "grad_norm": 1.6758607625961304, | |
| "learning_rate": 1.6441889790040297e-05, | |
| "loss": 0.325, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.8356643356643356, | |
| "grad_norm": 1.697481632232666, | |
| "learning_rate": 1.5963424501779844e-05, | |
| "loss": 0.3411, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 2.4904088973999023, | |
| "learning_rate": 1.5462757647453973e-05, | |
| "loss": 0.3325, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_f1": 0.8353484993816125, | |
| "eval_loss": 0.36182957887649536, | |
| "eval_runtime": 34.3605, | |
| "eval_samples_per_second": 1064.506, | |
| "eval_steps_per_second": 8.324, | |
| "step": 2288 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 5720, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.35103313039364e+16, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |