| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 201, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14925373134328357, | |
| "grad_norm": 19.874767303466797, | |
| "learning_rate": 4.7512437810945275e-05, | |
| "loss": 0.4626, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.29850746268656714, | |
| "grad_norm": 2.962965488433838, | |
| "learning_rate": 4.502487562189055e-05, | |
| "loss": 0.0398, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.44776119402985076, | |
| "grad_norm": 7.356692790985107, | |
| "learning_rate": 4.253731343283582e-05, | |
| "loss": 0.0239, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.5970149253731343, | |
| "grad_norm": 6.064937591552734, | |
| "learning_rate": 4.00497512437811e-05, | |
| "loss": 0.0357, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.746268656716418, | |
| "grad_norm": 9.119353294372559, | |
| "learning_rate": 3.756218905472637e-05, | |
| "loss": 0.0316, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.8955223880597015, | |
| "grad_norm": 7.955826282501221, | |
| "learning_rate": 3.5074626865671645e-05, | |
| "loss": 0.0496, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.05542003735899925, | |
| "eval_mse": 0.05542004108428955, | |
| "eval_runtime": 13.3429, | |
| "eval_samples_per_second": 10.043, | |
| "eval_steps_per_second": 1.274, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 1.044776119402985, | |
| "grad_norm": 0.8896350860595703, | |
| "learning_rate": 3.258706467661692e-05, | |
| "loss": 0.0319, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.1940298507462686, | |
| "grad_norm": 6.134321212768555, | |
| "learning_rate": 3.009950248756219e-05, | |
| "loss": 0.0244, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.3432835820895521, | |
| "grad_norm": 2.1838057041168213, | |
| "learning_rate": 2.7611940298507467e-05, | |
| "loss": 0.0174, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.4925373134328357, | |
| "grad_norm": 4.814957618713379, | |
| "learning_rate": 2.512437810945274e-05, | |
| "loss": 0.0127, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.6417910447761193, | |
| "grad_norm": 0.4210551381111145, | |
| "learning_rate": 2.2636815920398012e-05, | |
| "loss": 0.0318, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.7910447761194028, | |
| "grad_norm": 1.4994094371795654, | |
| "learning_rate": 2.0149253731343285e-05, | |
| "loss": 0.0068, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.9402985074626866, | |
| "grad_norm": 5.756369113922119, | |
| "learning_rate": 1.7661691542288558e-05, | |
| "loss": 0.0228, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.015796519815921783, | |
| "eval_mse": 0.015796517953276634, | |
| "eval_runtime": 13.4371, | |
| "eval_samples_per_second": 9.972, | |
| "eval_steps_per_second": 1.265, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 2.08955223880597, | |
| "grad_norm": 3.0607242584228516, | |
| "learning_rate": 1.5174129353233832e-05, | |
| "loss": 0.01, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.2388059701492535, | |
| "grad_norm": 1.876637578010559, | |
| "learning_rate": 1.2686567164179105e-05, | |
| "loss": 0.0061, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.388059701492537, | |
| "grad_norm": 0.6131510734558105, | |
| "learning_rate": 1.0199004975124378e-05, | |
| "loss": 0.0124, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.5373134328358207, | |
| "grad_norm": 2.405625104904175, | |
| "learning_rate": 7.711442786069652e-06, | |
| "loss": 0.007, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.6865671641791042, | |
| "grad_norm": 2.386357069015503, | |
| "learning_rate": 5.2238805970149255e-06, | |
| "loss": 0.0065, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.835820895522388, | |
| "grad_norm": 2.0660793781280518, | |
| "learning_rate": 2.736318407960199e-06, | |
| "loss": 0.0047, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.9850746268656714, | |
| "grad_norm": 2.1462905406951904, | |
| "learning_rate": 2.4875621890547267e-07, | |
| "loss": 0.0069, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 201, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |