| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9956331877729258, | |
| "eval_steps": 500, | |
| "global_step": 114, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.008733624454148471, | |
| "grad_norm": 648.0, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 56.9969, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.043668122270742356, | |
| "grad_norm": 394.0, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 51.7017, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.08733624454148471, | |
| "grad_norm": 30.25, | |
| "learning_rate": 0.0001666666666666667, | |
| "loss": 22.1529, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.13100436681222707, | |
| "grad_norm": 8.8125, | |
| "learning_rate": 0.00019957341762950344, | |
| "loss": 17.6564, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.17467248908296942, | |
| "grad_norm": 23.0, | |
| "learning_rate": 0.00019697969360350098, | |
| "loss": 15.7963, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2183406113537118, | |
| "grad_norm": 108.5, | |
| "learning_rate": 0.0001920905517944954, | |
| "loss": 8.7143, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.26200873362445415, | |
| "grad_norm": 7.9375, | |
| "learning_rate": 0.00018502171357296144, | |
| "loss": 2.0425, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3056768558951965, | |
| "grad_norm": 22.5, | |
| "learning_rate": 0.00017594049166547073, | |
| "loss": 1.6476, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.34934497816593885, | |
| "grad_norm": 1.4765625, | |
| "learning_rate": 0.0001650618300204242, | |
| "loss": 1.3805, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3930131004366812, | |
| "grad_norm": 2.109375, | |
| "learning_rate": 0.0001526432162877356, | |
| "loss": 1.2737, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.4366812227074236, | |
| "grad_norm": 3.0625, | |
| "learning_rate": 0.00013897858732926793, | |
| "loss": 1.2169, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.48034934497816595, | |
| "grad_norm": 1.7734375, | |
| "learning_rate": 0.00012439137201083773, | |
| "loss": 1.1317, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.5240174672489083, | |
| "grad_norm": 1.515625, | |
| "learning_rate": 0.00010922683594633021, | |
| "loss": 1.0869, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5676855895196506, | |
| "grad_norm": 1.7578125, | |
| "learning_rate": 9.384390938660572e-05, | |
| "loss": 1.0091, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.611353711790393, | |
| "grad_norm": 2.5625, | |
| "learning_rate": 7.860669167935028e-05, | |
| "loss": 1.0185, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6550218340611353, | |
| "grad_norm": 1.28125, | |
| "learning_rate": 6.387583338128471e-05, | |
| "loss": 0.9505, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.6986899563318777, | |
| "grad_norm": 1.6796875, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 0.9331, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.74235807860262, | |
| "grad_norm": 1.7734375, | |
| "learning_rate": 3.7307619410589376e-05, | |
| "loss": 0.9083, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.7860262008733624, | |
| "grad_norm": 0.9921875, | |
| "learning_rate": 2.6099108277934103e-05, | |
| "loss": 0.884, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8296943231441049, | |
| "grad_norm": 1.328125, | |
| "learning_rate": 1.663976147788806e-05, | |
| "loss": 0.9173, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.8733624454148472, | |
| "grad_norm": 0.85546875, | |
| "learning_rate": 9.153472818047625e-06, | |
| "loss": 0.8598, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.9170305676855895, | |
| "grad_norm": 0.69140625, | |
| "learning_rate": 3.817435682718096e-06, | |
| "loss": 0.8639, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.9606986899563319, | |
| "grad_norm": 0.71875, | |
| "learning_rate": 7.579490328064265e-07, | |
| "loss": 0.8678, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.9956331877729258, | |
| "eval_loss": 2.2895236015319824, | |
| "eval_runtime": 1.0009, | |
| "eval_samples_per_second": 4.996, | |
| "eval_steps_per_second": 1.998, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 0.9956331877729258, | |
| "step": 114, | |
| "total_flos": 1.742915407875932e+17, | |
| "train_loss": 5.99931854532476, | |
| "train_runtime": 898.0288, | |
| "train_samples_per_second": 2.039, | |
| "train_steps_per_second": 0.127 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 114, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.742915407875932e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |