| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0002732643614085138, | |
| "eval_steps": 25, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 5.465287228170276e-06, | |
| "eval_loss": 11.089583396911621, | |
| "eval_runtime": 568.5354, | |
| "eval_samples_per_second": 35.76, | |
| "eval_steps_per_second": 35.76, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 2.7326436140851382e-05, | |
| "grad_norm": 3.073105573654175, | |
| "learning_rate": 5e-05, | |
| "loss": 11.0125, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 5.4652872281702765e-05, | |
| "grad_norm": 3.04451322555542, | |
| "learning_rate": 0.0001, | |
| "loss": 11.0694, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 8.197930842255415e-05, | |
| "grad_norm": 2.6404738426208496, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 11.1158, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00010930574456340553, | |
| "grad_norm": 4.368824481964111, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 11.0983, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0001366321807042569, | |
| "grad_norm": 3.3952717781066895, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 11.1042, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0001366321807042569, | |
| "eval_loss": 11.052066802978516, | |
| "eval_runtime": 569.8381, | |
| "eval_samples_per_second": 35.679, | |
| "eval_steps_per_second": 35.679, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0001639586168451083, | |
| "grad_norm": 2.305196762084961, | |
| "learning_rate": 5e-05, | |
| "loss": 11.0493, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.00019128505298595966, | |
| "grad_norm": 2.923957586288452, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 11.0813, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.00021861148912681106, | |
| "grad_norm": 3.5035717487335205, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 10.901, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.00024593792526766246, | |
| "grad_norm": 3.045618772506714, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 10.987, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0002732643614085138, | |
| "grad_norm": 2.9671630859375, | |
| "learning_rate": 0.0, | |
| "loss": 10.9988, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0002732643614085138, | |
| "eval_loss": 11.035735130310059, | |
| "eval_runtime": 569.5957, | |
| "eval_samples_per_second": 35.694, | |
| "eval_steps_per_second": 35.694, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 259994419200.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |