| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.8622754491017965, |
| "eval_steps": 500, |
| "global_step": 20, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.2874251497005988, |
| "grad_norm": 10.670862197875977, |
| "learning_rate": 4.72751631047092e-05, |
| "loss": 1.0435, |
| "num_input_tokens_seen": 670576, |
| "step": 3 |
| }, |
| { |
| "epoch": 0.5748502994011976, |
| "grad_norm": 1.6404894590377808, |
| "learning_rate": 3.969463130731183e-05, |
| "loss": 0.6701, |
| "num_input_tokens_seen": 1356912, |
| "step": 6 |
| }, |
| { |
| "epoch": 0.8622754491017964, |
| "grad_norm": 1.251344919204712, |
| "learning_rate": 2.8910861626005776e-05, |
| "loss": 0.4917, |
| "num_input_tokens_seen": 2040576, |
| "step": 9 |
| }, |
| { |
| "epoch": 1.095808383233533, |
| "grad_norm": 1.4765971899032593, |
| "learning_rate": 1.7274575140626318e-05, |
| "loss": 0.4176, |
| "num_input_tokens_seen": 2590528, |
| "step": 12 |
| }, |
| { |
| "epoch": 1.3832335329341316, |
| "grad_norm": 0.6158081293106079, |
| "learning_rate": 7.3223304703363135e-06, |
| "loss": 0.3634, |
| "num_input_tokens_seen": 3275616, |
| "step": 15 |
| }, |
| { |
| "epoch": 1.6706586826347305, |
| "grad_norm": 0.5957555174827576, |
| "learning_rate": 1.2235870926211619e-06, |
| "loss": 0.3434, |
| "num_input_tokens_seen": 3959408, |
| "step": 18 |
| }, |
| { |
| "epoch": 1.8622754491017965, |
| "num_input_tokens_seen": 4416112, |
| "step": 20, |
| "total_flos": 6389928689664.0, |
| "train_loss": 0.5334016799926757, |
| "train_runtime": 2199.0599, |
| "train_samples_per_second": 3.638, |
| "train_steps_per_second": 0.009 |
| } |
| ], |
| "logging_steps": 3, |
| "max_steps": 20, |
| "num_input_tokens_seen": 4416112, |
| "num_train_epochs": 2, |
| "save_steps": 30, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 6389928689664.0, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
