| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.4705882352941178, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.014705882352941176, | |
| "eval_loss": NaN, | |
| "eval_runtime": 17.0771, | |
| "eval_samples_per_second": 6.734, | |
| "eval_steps_per_second": 0.878, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.04411764705882353, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 2.2991, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.08823529411764706, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 2.5431, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.1323529411764706, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 13.2201, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.1323529411764706, | |
| "eval_loss": NaN, | |
| "eval_runtime": 16.7373, | |
| "eval_samples_per_second": 6.871, | |
| "eval_steps_per_second": 0.896, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.17647058823529413, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 16.8201, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.22058823529411764, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 1.7364, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2647058823529412, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 3.6238, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.2647058823529412, | |
| "eval_loss": NaN, | |
| "eval_runtime": 16.7549, | |
| "eval_samples_per_second": 6.864, | |
| "eval_steps_per_second": 0.895, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.3088235294117647, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 22.9854, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.35294117647058826, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.8414, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.39705882352941174, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 15.9645, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.39705882352941174, | |
| "eval_loss": NaN, | |
| "eval_runtime": 16.7723, | |
| "eval_samples_per_second": 6.857, | |
| "eval_steps_per_second": 0.894, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.4411764705882353, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 3.1853, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4852941176470588, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 8.3073, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.5294117647058824, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 1.5029, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.5294117647058824, | |
| "eval_loss": NaN, | |
| "eval_runtime": 16.7719, | |
| "eval_samples_per_second": 6.857, | |
| "eval_steps_per_second": 0.894, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.5735294117647058, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 4.2828, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.6176470588235294, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.9355, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.6617647058823529, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 10.799, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6617647058823529, | |
| "eval_loss": NaN, | |
| "eval_runtime": 16.7722, | |
| "eval_samples_per_second": 6.857, | |
| "eval_steps_per_second": 0.894, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 15.3127, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 1.8674, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.7941176470588235, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 16.1705, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.7941176470588235, | |
| "eval_loss": NaN, | |
| "eval_runtime": 16.7643, | |
| "eval_samples_per_second": 6.86, | |
| "eval_steps_per_second": 0.895, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.8382352941176471, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.0, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.8823529411764706, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 22.9755, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.9264705882352942, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 8.2233, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.9264705882352942, | |
| "eval_loss": NaN, | |
| "eval_runtime": 16.7523, | |
| "eval_samples_per_second": 6.865, | |
| "eval_steps_per_second": 0.895, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.9705882352941176, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 6.5478, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.0147058823529411, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 3.7818, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.0588235294117647, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 17.1026, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.0588235294117647, | |
| "eval_loss": NaN, | |
| "eval_runtime": 16.7798, | |
| "eval_samples_per_second": 6.853, | |
| "eval_steps_per_second": 0.894, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.1029411764705883, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 2.6388, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.1470588235294117, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 2.2662, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.1911764705882353, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 9.0086, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.1911764705882353, | |
| "eval_loss": NaN, | |
| "eval_runtime": 16.8654, | |
| "eval_samples_per_second": 6.819, | |
| "eval_steps_per_second": 0.889, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.2352941176470589, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 15.8322, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.2794117647058822, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 22.2182, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.3235294117647058, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 8.1206, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.3235294117647058, | |
| "eval_loss": NaN, | |
| "eval_runtime": 16.849, | |
| "eval_samples_per_second": 6.825, | |
| "eval_steps_per_second": 0.89, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.3676470588235294, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 2.6242, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 3.3084, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.4558823529411764, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 2.4601, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.4558823529411764, | |
| "eval_loss": NaN, | |
| "eval_runtime": 16.8479, | |
| "eval_samples_per_second": 6.826, | |
| "eval_steps_per_second": 0.89, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.8375566076870656e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |