| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0002460732855459013, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 4.921465710918026e-06, | |
| "eval_loss": 1.5265452861785889, | |
| "eval_runtime": 1158.1597, | |
| "eval_samples_per_second": 36.936, | |
| "eval_steps_per_second": 18.468, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 4.9214657109180254e-05, | |
| "grad_norm": 1.8430445194244385, | |
| "learning_rate": 0.0002, | |
| "loss": 1.117, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.397905424193433e-05, | |
| "eval_loss": 0.7989104986190796, | |
| "eval_runtime": 1141.9931, | |
| "eval_samples_per_second": 37.459, | |
| "eval_steps_per_second": 18.73, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 9.842931421836051e-05, | |
| "grad_norm": 3.3933048248291016, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7535, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00012795810848386865, | |
| "eval_loss": 0.6970112323760986, | |
| "eval_runtime": 1148.9757, | |
| "eval_samples_per_second": 37.231, | |
| "eval_steps_per_second": 18.616, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.00014764397132754076, | |
| "grad_norm": 3.2244198322296143, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7387, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.000191937162725803, | |
| "eval_loss": 0.6604138612747192, | |
| "eval_runtime": 1150.8851, | |
| "eval_samples_per_second": 37.17, | |
| "eval_steps_per_second": 18.585, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.00019685862843672102, | |
| "grad_norm": 3.082199811935425, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8019, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0002460732855459013, | |
| "grad_norm": 2.879859209060669, | |
| "learning_rate": 0.0002, | |
| "loss": 0.6401, | |
| "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": 1419983835365376.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |