| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.2552648372686662, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005105296745373325, | |
| "eval_loss": 0.6308530569076538, | |
| "eval_runtime": 34.3499, | |
| "eval_samples_per_second": 38.428, | |
| "eval_steps_per_second": 4.804, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05105296745373325, | |
| "grad_norm": 0.10898499190807343, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.6061, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05105296745373325, | |
| "eval_loss": 0.6026803851127625, | |
| "eval_runtime": 34.5662, | |
| "eval_samples_per_second": 38.188, | |
| "eval_steps_per_second": 4.773, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1021059349074665, | |
| "grad_norm": 0.06684104353189468, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.5841, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1021059349074665, | |
| "eval_loss": 0.5628232359886169, | |
| "eval_runtime": 34.3843, | |
| "eval_samples_per_second": 38.39, | |
| "eval_steps_per_second": 4.799, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.15315890236119975, | |
| "grad_norm": 0.07069219648838043, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 0.5459, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.15315890236119975, | |
| "eval_loss": 0.5473927855491638, | |
| "eval_runtime": 34.3812, | |
| "eval_samples_per_second": 38.393, | |
| "eval_steps_per_second": 4.799, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.204211869814933, | |
| "grad_norm": 0.0595882311463356, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 0.5444, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.204211869814933, | |
| "eval_loss": 0.540993332862854, | |
| "eval_runtime": 34.4989, | |
| "eval_samples_per_second": 38.262, | |
| "eval_steps_per_second": 4.783, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2552648372686662, | |
| "grad_norm": 0.05125812813639641, | |
| "learning_rate": 0.0, | |
| "loss": 0.5309, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2552648372686662, | |
| "eval_loss": 0.5400261282920837, | |
| "eval_runtime": 34.4315, | |
| "eval_samples_per_second": 38.337, | |
| "eval_steps_per_second": 4.792, | |
| "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": 4.09267165200384e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |