| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0016304569355561896, | |
| "eval_steps": 3, | |
| "global_step": 10, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00016304569355561895, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2121, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00016304569355561895, | |
| "eval_loss": NaN, | |
| "eval_runtime": 41.549, | |
| "eval_samples_per_second": 62.168, | |
| "eval_steps_per_second": 31.096, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0003260913871112379, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0004891370806668569, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0004891370806668569, | |
| "eval_loss": NaN, | |
| "eval_runtime": 40.5147, | |
| "eval_samples_per_second": 63.755, | |
| "eval_steps_per_second": 31.89, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0006521827742224758, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0008152284677780948, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0009782741613337139, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012, | |
| "loss": 0.2186, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0009782741613337139, | |
| "eval_loss": NaN, | |
| "eval_runtime": 40.3666, | |
| "eval_samples_per_second": 63.988, | |
| "eval_steps_per_second": 32.007, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0011413198548893327, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014, | |
| "loss": 0.211, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0013043655484449516, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0014674112420005707, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018, | |
| "loss": 2.347, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0014674112420005707, | |
| "eval_loss": NaN, | |
| "eval_runtime": 40.0558, | |
| "eval_samples_per_second": 64.485, | |
| "eval_steps_per_second": 32.255, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0016304569355561896, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0002, | |
| "loss": 0.6112, | |
| "step": 10 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 10, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 3, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 384872824700928.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |