| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 102, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.029411764705882353, | |
| "grad_norm": 8.486539952545904, | |
| "learning_rate": 1.8181818181818183e-06, | |
| "loss": 1.6655, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.14705882352941177, | |
| "grad_norm": 3.662688033003511, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 1.5462, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.29411764705882354, | |
| "grad_norm": 2.308882684234435, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 1.3597, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4411764705882353, | |
| "grad_norm": 2.6240427599001044, | |
| "learning_rate": 1.9904804439875635e-05, | |
| "loss": 1.3014, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5882352941176471, | |
| "grad_norm": 1.7634831845318764, | |
| "learning_rate": 1.952117665910714e-05, | |
| "loss": 1.2951, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7352941176470589, | |
| "grad_norm": 1.6277047419122932, | |
| "learning_rate": 1.8854560256532098e-05, | |
| "loss": 1.2399, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.8823529411764706, | |
| "grad_norm": 1.5908199398295786, | |
| "learning_rate": 1.7924768419510906e-05, | |
| "loss": 1.171, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.8685240745544434, | |
| "eval_runtime": 0.2849, | |
| "eval_samples_per_second": 10.531, | |
| "eval_steps_per_second": 3.51, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.0294117647058822, | |
| "grad_norm": 1.5592035755033316, | |
| "learning_rate": 1.6759436441447544e-05, | |
| "loss": 1.1548, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 1.3281842310864844, | |
| "learning_rate": 1.5393200344991993e-05, | |
| "loss": 0.8484, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.3235294117647058, | |
| "grad_norm": 1.3525623421140405, | |
| "learning_rate": 1.3866667429414188e-05, | |
| "loss": 0.757, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.4705882352941178, | |
| "grad_norm": 1.3499111513522848, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.772, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.6176470588235294, | |
| "grad_norm": 1.3440352220669844, | |
| "learning_rate": 1.0517613528842096e-05, | |
| "loss": 0.7721, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.7647058823529411, | |
| "grad_norm": 1.243123047867263, | |
| "learning_rate": 8.79463319744677e-06, | |
| "loss": 0.7588, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.9117647058823528, | |
| "grad_norm": 1.3447383678590699, | |
| "learning_rate": 7.107478804634324e-06, | |
| "loss": 0.7274, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.986986756324768, | |
| "eval_runtime": 0.2954, | |
| "eval_samples_per_second": 10.157, | |
| "eval_steps_per_second": 3.386, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.0588235294117645, | |
| "grad_norm": 1.6313000577136987, | |
| "learning_rate": 5.5062959903283855e-06, | |
| "loss": 0.5966, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.2058823529411766, | |
| "grad_norm": 1.2766661386646334, | |
| "learning_rate": 4.038675145307747e-06, | |
| "loss": 0.4519, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 1.1360520079297232, | |
| "learning_rate": 2.748236928566238e-06, | |
| "loss": 0.468, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 1.3830749450797224, | |
| "learning_rate": 1.6733357731279375e-06, | |
| "loss": 0.4451, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.6470588235294117, | |
| "grad_norm": 1.2190766481868696, | |
| "learning_rate": 8.459199147463371e-07, | |
| "loss": 0.4314, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.7941176470588234, | |
| "grad_norm": 1.3481983243287725, | |
| "learning_rate": 2.905818257394799e-07, | |
| "loss": 0.4465, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "grad_norm": 1.2131662765102487, | |
| "learning_rate": 2.3827276987524738e-08, | |
| "loss": 0.4153, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.1628994941711426, | |
| "eval_runtime": 0.2916, | |
| "eval_samples_per_second": 10.289, | |
| "eval_steps_per_second": 3.43, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 102, | |
| "total_flos": 6684244180992.0, | |
| "train_loss": 0.8398442899479586, | |
| "train_runtime": 366.4522, | |
| "train_samples_per_second": 8.735, | |
| "train_steps_per_second": 0.278 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 102, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6684244180992.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |