| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.10906612133605999, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0021813224267212, | |
| "eval_loss": 1.8513853549957275, | |
| "eval_runtime": 82.0055, | |
| "eval_samples_per_second": 37.668, | |
| "eval_steps_per_second": 4.719, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.021813224267211998, | |
| "grad_norm": 0.24979163706302643, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 1.8294, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.021813224267211998, | |
| "eval_loss": 1.797348141670227, | |
| "eval_runtime": 81.9634, | |
| "eval_samples_per_second": 37.688, | |
| "eval_steps_per_second": 4.722, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.043626448534423996, | |
| "grad_norm": 0.22572024166584015, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 1.8118, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.043626448534423996, | |
| "eval_loss": 1.7619872093200684, | |
| "eval_runtime": 82.0518, | |
| "eval_samples_per_second": 37.647, | |
| "eval_steps_per_second": 4.717, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.065439672801636, | |
| "grad_norm": 0.20880632102489471, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 1.7614, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.065439672801636, | |
| "eval_loss": 1.7445645332336426, | |
| "eval_runtime": 81.8885, | |
| "eval_samples_per_second": 37.722, | |
| "eval_steps_per_second": 4.726, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.08725289706884799, | |
| "grad_norm": 0.222714364528656, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 1.7566, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.08725289706884799, | |
| "eval_loss": 1.7374491691589355, | |
| "eval_runtime": 81.9879, | |
| "eval_samples_per_second": 37.676, | |
| "eval_steps_per_second": 4.72, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10906612133605999, | |
| "grad_norm": 0.21956880390644073, | |
| "learning_rate": 0.0, | |
| "loss": 1.7245, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.10906612133605999, | |
| "eval_loss": 1.736304759979248, | |
| "eval_runtime": 82.283, | |
| "eval_samples_per_second": 37.541, | |
| "eval_steps_per_second": 4.703, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.843917336700518e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |