| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.04008016032064128, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.002004008016032064, | |
| "grad_norm": 0.9981935024261475, | |
| "learning_rate": 1e-05, | |
| "loss": 3.6274, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.002004008016032064, | |
| "eval_loss": 3.705083131790161, | |
| "eval_runtime": 67.9972, | |
| "eval_samples_per_second": 24.722, | |
| "eval_steps_per_second": 12.368, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004008016032064128, | |
| "grad_norm": 1.0312002897262573, | |
| "learning_rate": 2e-05, | |
| "loss": 3.7344, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.006012024048096192, | |
| "grad_norm": 1.0591121912002563, | |
| "learning_rate": 3e-05, | |
| "loss": 3.6015, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.008016032064128256, | |
| "grad_norm": 1.0240955352783203, | |
| "learning_rate": 4e-05, | |
| "loss": 3.7196, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.01002004008016032, | |
| "grad_norm": 1.0419278144836426, | |
| "learning_rate": 5e-05, | |
| "loss": 3.7358, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01002004008016032, | |
| "eval_loss": 3.6875150203704834, | |
| "eval_runtime": 66.7319, | |
| "eval_samples_per_second": 25.19, | |
| "eval_steps_per_second": 12.603, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.012024048096192385, | |
| "grad_norm": 1.075563669204712, | |
| "learning_rate": 6e-05, | |
| "loss": 3.6583, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.014028056112224449, | |
| "grad_norm": 1.0435458421707153, | |
| "learning_rate": 7e-05, | |
| "loss": 3.6918, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.01603206412825651, | |
| "grad_norm": 0.9885270595550537, | |
| "learning_rate": 8e-05, | |
| "loss": 3.6284, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.018036072144288578, | |
| "grad_norm": 1.0842863321304321, | |
| "learning_rate": 9e-05, | |
| "loss": 3.6854, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.02004008016032064, | |
| "grad_norm": 1.0496262311935425, | |
| "learning_rate": 0.0001, | |
| "loss": 3.5911, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02004008016032064, | |
| "eval_loss": 3.535064935684204, | |
| "eval_runtime": 66.7323, | |
| "eval_samples_per_second": 25.19, | |
| "eval_steps_per_second": 12.603, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.022044088176352707, | |
| "grad_norm": 1.0535022020339966, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 3.4243, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.02404809619238477, | |
| "grad_norm": 1.0760126113891602, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 3.4287, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.026052104208416832, | |
| "grad_norm": 1.047857403755188, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 3.4245, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.028056112224448898, | |
| "grad_norm": 0.9840763807296753, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 3.2842, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.03006012024048096, | |
| "grad_norm": 0.9731857180595398, | |
| "learning_rate": 5e-05, | |
| "loss": 3.2958, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.03006012024048096, | |
| "eval_loss": 3.302708148956299, | |
| "eval_runtime": 66.7476, | |
| "eval_samples_per_second": 25.184, | |
| "eval_steps_per_second": 12.6, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.03206412825651302, | |
| "grad_norm": 0.9710604548454285, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 3.2329, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.03406813627254509, | |
| "grad_norm": 0.9070257544517517, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 3.1751, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.036072144288577156, | |
| "grad_norm": 0.9331132769584656, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 3.1826, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.03807615230460922, | |
| "grad_norm": 0.9252903461456299, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 3.1469, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.04008016032064128, | |
| "grad_norm": 0.9361170530319214, | |
| "learning_rate": 0.0, | |
| "loss": 3.2011, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.04008016032064128, | |
| "eval_loss": 3.241048812866211, | |
| "eval_runtime": 66.7398, | |
| "eval_samples_per_second": 25.187, | |
| "eval_steps_per_second": 12.601, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4442802754682880.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |