| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.001796541657309679, | |
| "eval_steps": 3, | |
| "global_step": 6, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0002994236095516131, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 10.9802, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0002994236095516131, | |
| "eval_loss": 11.5775728225708, | |
| "eval_runtime": 95.8713, | |
| "eval_samples_per_second": 7.343, | |
| "eval_steps_per_second": 7.343, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0005988472191032262, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 10.7165, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0008982708286548395, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 10.434, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0008982708286548395, | |
| "eval_loss": 11.5775728225708, | |
| "eval_runtime": 94.6943, | |
| "eval_samples_per_second": 7.434, | |
| "eval_steps_per_second": 7.434, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0011976944382064525, | |
| "grad_norm": 11.335941314697266, | |
| "learning_rate": 2e-05, | |
| "loss": 9.5838, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0014971180477580657, | |
| "grad_norm": 12.509073257446289, | |
| "learning_rate": 4e-05, | |
| "loss": 11.0567, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.001796541657309679, | |
| "grad_norm": 13.569826126098633, | |
| "learning_rate": 6e-05, | |
| "loss": 12.2136, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.001796541657309679, | |
| "eval_loss": 11.521071434020996, | |
| "eval_runtime": 94.7717, | |
| "eval_samples_per_second": 7.428, | |
| "eval_steps_per_second": 7.428, | |
| "step": 6 | |
| } | |
| ], | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1109738548887552.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |