| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.6915769474350855, | |
| "eval_steps": 500, | |
| "global_step": 8500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15832805573147563, | |
| "grad_norm": 0.43304896354675293, | |
| "learning_rate": 4.7361199071142074e-05, | |
| "loss": 0.0026, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.31665611146295125, | |
| "grad_norm": 0.2186095416545868, | |
| "learning_rate": 4.472239814228415e-05, | |
| "loss": 0.0007, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.4749841671944269, | |
| "grad_norm": 0.22537918388843536, | |
| "learning_rate": 4.2083597213426224e-05, | |
| "loss": 0.0005, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.6333122229259025, | |
| "grad_norm": 0.046680252999067307, | |
| "learning_rate": 3.944479628456829e-05, | |
| "loss": 0.0004, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.7916402786573781, | |
| "grad_norm": 0.15091869235038757, | |
| "learning_rate": 3.680599535571037e-05, | |
| "loss": 0.0003, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.9499683343888538, | |
| "grad_norm": 0.1729196012020111, | |
| "learning_rate": 3.416719442685244e-05, | |
| "loss": 0.0002, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.0001189145477837883, | |
| "eval_runtime": 134.5836, | |
| "eval_samples_per_second": 46.93, | |
| "eval_steps_per_second": 5.87, | |
| "step": 3158 | |
| }, | |
| { | |
| "epoch": 1.1082963901203293, | |
| "grad_norm": 0.13886131346225739, | |
| "learning_rate": 3.152839349799451e-05, | |
| "loss": 0.0002, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.266624445851805, | |
| "grad_norm": 0.0958346351981163, | |
| "learning_rate": 2.8889592569136586e-05, | |
| "loss": 0.0002, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.4249525015832805, | |
| "grad_norm": 0.04034944623708725, | |
| "learning_rate": 2.625079164027866e-05, | |
| "loss": 0.0001, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.5832805573147561, | |
| "grad_norm": 0.08302062004804611, | |
| "learning_rate": 2.3611990711420732e-05, | |
| "loss": 0.0001, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.7416086130462318, | |
| "grad_norm": 0.016813725233078003, | |
| "learning_rate": 2.0973189782562804e-05, | |
| "loss": 0.0001, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.8999366687777073, | |
| "grad_norm": 0.027263302356004715, | |
| "learning_rate": 1.833438885370488e-05, | |
| "loss": 0.0001, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.1332004760333803e-05, | |
| "eval_runtime": 134.3012, | |
| "eval_samples_per_second": 47.029, | |
| "eval_steps_per_second": 5.882, | |
| "step": 6316 | |
| }, | |
| { | |
| "epoch": 2.058264724509183, | |
| "grad_norm": 0.008972160518169403, | |
| "learning_rate": 1.569558792484695e-05, | |
| "loss": 0.0001, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.2165927802406586, | |
| "grad_norm": 0.028740577399730682, | |
| "learning_rate": 1.3056786995989024e-05, | |
| "loss": 0.0, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.374920835972134, | |
| "grad_norm": 0.014961929991841316, | |
| "learning_rate": 1.0417986067131096e-05, | |
| "loss": 0.0, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.53324889170361, | |
| "grad_norm": 0.02067851461470127, | |
| "learning_rate": 7.779185138273169e-06, | |
| "loss": 0.0, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.6915769474350855, | |
| "grad_norm": 0.014241858385503292, | |
| "learning_rate": 5.140384209415242e-06, | |
| "loss": 0.0, | |
| "step": 8500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 9474, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 1.7891391123456e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |