| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 2.9937106918238996, |
| "eval_steps": 500, |
| "global_step": 357, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.4192872117400419, |
| "grad_norm": 0.057723674923181534, |
| "learning_rate": 0.00021813031161473087, |
| "loss": 11.9272, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.8385744234800838, |
| "grad_norm": 0.04375343769788742, |
| "learning_rate": 0.00018271954674220964, |
| "loss": 11.9208, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.9979035639412998, |
| "eval_loss": 11.919021606445312, |
| "eval_runtime": 0.4164, |
| "eval_samples_per_second": 242.533, |
| "eval_steps_per_second": 62.434, |
| "step": 119 |
| }, |
| { |
| "epoch": 1.2578616352201257, |
| "grad_norm": 0.028496425598859787, |
| "learning_rate": 0.00014730878186968837, |
| "loss": 12.1057, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.6771488469601676, |
| "grad_norm": 0.027717996388673782, |
| "learning_rate": 0.00011189801699716715, |
| "loss": 11.9206, |
| "step": 200 |
| }, |
| { |
| "epoch": 1.9958071278825997, |
| "eval_loss": 11.916767120361328, |
| "eval_runtime": 0.4141, |
| "eval_samples_per_second": 243.91, |
| "eval_steps_per_second": 62.789, |
| "step": 238 |
| }, |
| { |
| "epoch": 2.0964360587002098, |
| "grad_norm": 0.03638681024312973, |
| "learning_rate": 7.64872521246459e-05, |
| "loss": 12.0807, |
| "step": 250 |
| }, |
| { |
| "epoch": 2.5157232704402515, |
| "grad_norm": 0.029723290354013443, |
| "learning_rate": 4.107648725212465e-05, |
| "loss": 11.927, |
| "step": 300 |
| }, |
| { |
| "epoch": 2.9350104821802936, |
| "grad_norm": 0.08462727814912796, |
| "learning_rate": 5.6657223796034e-06, |
| "loss": 11.916, |
| "step": 350 |
| }, |
| { |
| "epoch": 2.9937106918238996, |
| "eval_loss": 11.916586875915527, |
| "eval_runtime": 0.4094, |
| "eval_samples_per_second": 246.686, |
| "eval_steps_per_second": 63.503, |
| "step": 357 |
| } |
| ], |
| "logging_steps": 50, |
| "max_steps": 358, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 4, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 402378350592.0, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
