| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.12746972594008923, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0012746972594008922, | |
| "eval_loss": NaN, | |
| "eval_runtime": 90.7061, | |
| "eval_samples_per_second": 14.564, | |
| "eval_steps_per_second": 1.83, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0038240917782026767, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0076481835564053535, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.011472275334608031, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.011472275334608031, | |
| "eval_loss": NaN, | |
| "eval_runtime": 91.231, | |
| "eval_samples_per_second": 14.48, | |
| "eval_steps_per_second": 1.82, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.015296367112810707, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.019120458891013385, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.022944550669216062, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.022944550669216062, | |
| "eval_loss": NaN, | |
| "eval_runtime": 91.264, | |
| "eval_samples_per_second": 14.474, | |
| "eval_steps_per_second": 1.819, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.02676864244741874, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.0, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.030592734225621414, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.03441682600382409, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.0, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.03441682600382409, | |
| "eval_loss": NaN, | |
| "eval_runtime": 91.2568, | |
| "eval_samples_per_second": 14.476, | |
| "eval_steps_per_second": 1.819, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.03824091778202677, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04206500956022945, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.0, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.045889101338432124, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.045889101338432124, | |
| "eval_loss": NaN, | |
| "eval_runtime": 91.2737, | |
| "eval_samples_per_second": 14.473, | |
| "eval_steps_per_second": 1.819, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0497131931166348, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.0, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.05353728489483748, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.05736137667304015, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.05736137667304015, | |
| "eval_loss": NaN, | |
| "eval_runtime": 91.2609, | |
| "eval_samples_per_second": 14.475, | |
| "eval_steps_per_second": 1.819, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.06118546845124283, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.0, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.06500956022944551, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.0, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.06883365200764818, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.0, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.06883365200764818, | |
| "eval_loss": NaN, | |
| "eval_runtime": 91.2629, | |
| "eval_samples_per_second": 14.475, | |
| "eval_steps_per_second": 1.819, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.07265774378585087, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.0, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.07648183556405354, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.08030592734225621, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.0, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.08030592734225621, | |
| "eval_loss": NaN, | |
| "eval_runtime": 91.2492, | |
| "eval_samples_per_second": 14.477, | |
| "eval_steps_per_second": 1.819, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.0841300191204589, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.0, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.08795411089866156, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.0, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.09177820267686425, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.0, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.09177820267686425, | |
| "eval_loss": NaN, | |
| "eval_runtime": 91.3258, | |
| "eval_samples_per_second": 14.465, | |
| "eval_steps_per_second": 1.818, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.09560229445506692, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.0, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.0994263862332696, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.0, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.10325047801147227, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 0.0, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.10325047801147227, | |
| "eval_loss": NaN, | |
| "eval_runtime": 91.3147, | |
| "eval_samples_per_second": 14.466, | |
| "eval_steps_per_second": 1.818, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.10707456978967496, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.0, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.11089866156787763, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.0, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.1147227533460803, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1147227533460803, | |
| "eval_loss": NaN, | |
| "eval_runtime": 91.311, | |
| "eval_samples_per_second": 14.467, | |
| "eval_steps_per_second": 1.818, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.11854684512428298, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.0, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.12237093690248566, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.0, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.12619502868068833, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 0.0, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.12619502868068833, | |
| "eval_loss": NaN, | |
| "eval_runtime": 91.2841, | |
| "eval_samples_per_second": 14.471, | |
| "eval_steps_per_second": 1.818, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 1.3487508495846605e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |