| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9013539651837523, | |
| "eval_steps": 500, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.8388136686009026e-05, | |
| "loss": 0.6868, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 4.677627337201805e-05, | |
| "loss": 0.5552, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 4.516441005802708e-05, | |
| "loss": 0.5657, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 4.355254674403611e-05, | |
| "loss": 0.4919, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 4.1940683430045136e-05, | |
| "loss": 0.4673, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_loss": 0.4286854565143585, | |
| "eval_runtime": 157.876, | |
| "eval_samples_per_second": 13.093, | |
| "eval_steps_per_second": 1.641, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 4.032882011605416e-05, | |
| "loss": 0.4705, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 3.8716956802063184e-05, | |
| "loss": 0.4717, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 3.7105093488072215e-05, | |
| "loss": 0.4523, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 3.549323017408124e-05, | |
| "loss": 0.4333, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 3.388136686009026e-05, | |
| "loss": 0.4765, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_loss": 0.3951832354068756, | |
| "eval_runtime": 154.675, | |
| "eval_samples_per_second": 13.364, | |
| "eval_steps_per_second": 1.674, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 3.226950354609929e-05, | |
| "loss": 0.3246, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 3.065764023210832e-05, | |
| "loss": 0.3084, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 2.9045776918117345e-05, | |
| "loss": 0.3059, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 2.7433913604126372e-05, | |
| "loss": 0.3234, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 2.5822050290135396e-05, | |
| "loss": 0.3353, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "eval_loss": 0.5979052186012268, | |
| "eval_runtime": 152.068, | |
| "eval_samples_per_second": 13.593, | |
| "eval_steps_per_second": 1.703, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 2.4210186976144423e-05, | |
| "loss": 0.3539, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 2.259832366215345e-05, | |
| "loss": 0.3427, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 2.0986460348162475e-05, | |
| "loss": 0.2742, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 1.9374597034171506e-05, | |
| "loss": 0.2958, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 1.776273372018053e-05, | |
| "loss": 0.3052, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "eval_loss": 0.5818384885787964, | |
| "eval_runtime": 151.883, | |
| "eval_samples_per_second": 13.609, | |
| "eval_steps_per_second": 1.705, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 1.6150870406189557e-05, | |
| "loss": 0.3323, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 1.4539007092198581e-05, | |
| "loss": 0.1706, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 1.292714377820761e-05, | |
| "loss": 0.2092, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 1.1315280464216636e-05, | |
| "loss": 0.1928, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 9.703417150225661e-06, | |
| "loss": 0.2238, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "eval_loss": 0.6637986302375793, | |
| "eval_runtime": 152.045, | |
| "eval_samples_per_second": 13.595, | |
| "eval_steps_per_second": 1.703, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 8.091553836234687e-06, | |
| "loss": 0.1736, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 6.479690522243714e-06, | |
| "loss": 0.2078, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 4.867827208252741e-06, | |
| "loss": 0.1574, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 3.255963894261767e-06, | |
| "loss": 0.2115, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 1.644100580270793e-06, | |
| "loss": 0.147, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "eval_loss": 0.7667654156684875, | |
| "eval_runtime": 153.515, | |
| "eval_samples_per_second": 13.464, | |
| "eval_steps_per_second": 1.687, | |
| "step": 3000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 3102, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 1195734913300080.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |