| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.010826416456153013, | |
| "eval_steps": 8, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00036088054853843375, | |
| "eval_loss": NaN, | |
| "eval_runtime": 37.8067, | |
| "eval_samples_per_second": 30.868, | |
| "eval_steps_per_second": 15.447, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0010826416456153013, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 4.8267, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0021652832912306026, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00288704438830747, | |
| "eval_loss": NaN, | |
| "eval_runtime": 35.9434, | |
| "eval_samples_per_second": 32.468, | |
| "eval_steps_per_second": 16.248, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.003247924936845904, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.004330566582461205, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0054132082280765065, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00577408877661494, | |
| "eval_loss": NaN, | |
| "eval_runtime": 35.4561, | |
| "eval_samples_per_second": 32.914, | |
| "eval_steps_per_second": 16.471, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.006495849873691808, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.007578491519307109, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 0.0, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.00866113316492241, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.00866113316492241, | |
| "eval_loss": NaN, | |
| "eval_runtime": 36.192, | |
| "eval_samples_per_second": 32.245, | |
| "eval_steps_per_second": 16.136, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.009743774810537712, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 0.0, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.010826416456153013, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 30 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 2290345517776896.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |