| { |
| "best_global_step": 200, |
| "best_metric": 0.04782823845744133, |
| "best_model_checkpoint": "./lbtestcd_lora_model/checkpoint-200", |
| "epoch": 1.6285714285714286, |
| "eval_steps": 100, |
| "global_step": 200, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 1.6688840508460998, |
| "epoch": 0.08163265306122448, |
| "grad_norm": 1.984375, |
| "learning_rate": 0.00013846153846153847, |
| "loss": 2.7854, |
| "mean_token_accuracy": 0.5734112456440925, |
| "num_tokens": 29874.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 0.8716737445443868, |
| "epoch": 0.16326530612244897, |
| "grad_norm": 0.8046875, |
| "learning_rate": 0.00019484978540772535, |
| "loss": 0.7077, |
| "mean_token_accuracy": 0.8737192109227181, |
| "num_tokens": 59724.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 0.12969586309045553, |
| "epoch": 0.24489795918367346, |
| "grad_norm": 0.1884765625, |
| "learning_rate": 0.00018626609442060085, |
| "loss": 0.1144, |
| "mean_token_accuracy": 0.9771728187799453, |
| "num_tokens": 89501.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 0.10608638003468514, |
| "epoch": 0.32653061224489793, |
| "grad_norm": 0.119140625, |
| "learning_rate": 0.0001776824034334764, |
| "loss": 0.0903, |
| "mean_token_accuracy": 0.9806259959936142, |
| "num_tokens": 119257.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 0.09941115416586399, |
| "epoch": 0.40816326530612246, |
| "grad_norm": 0.11376953125, |
| "learning_rate": 0.00016909871244635193, |
| "loss": 0.0896, |
| "mean_token_accuracy": 0.9801410123705864, |
| "num_tokens": 149090.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.09959900509566069, |
| "epoch": 0.4897959183673469, |
| "grad_norm": 0.1416015625, |
| "learning_rate": 0.00016051502145922748, |
| "loss": 0.0788, |
| "mean_token_accuracy": 0.9814802333712578, |
| "num_tokens": 178953.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.10063424333930016, |
| "epoch": 0.5714285714285714, |
| "grad_norm": 0.1572265625, |
| "learning_rate": 0.000151931330472103, |
| "loss": 0.0671, |
| "mean_token_accuracy": 0.9840218782424927, |
| "num_tokens": 208876.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.06386051792651415, |
| "epoch": 0.6530612244897959, |
| "grad_norm": 0.099609375, |
| "learning_rate": 0.00014334763948497856, |
| "loss": 0.05, |
| "mean_token_accuracy": 0.987203674018383, |
| "num_tokens": 238734.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.05305444821715355, |
| "epoch": 0.7346938775510204, |
| "grad_norm": 0.06494140625, |
| "learning_rate": 0.00013476394849785408, |
| "loss": 0.0477, |
| "mean_token_accuracy": 0.9869530826807023, |
| "num_tokens": 268539.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.0516530342400074, |
| "epoch": 0.8163265306122449, |
| "grad_norm": 0.1142578125, |
| "learning_rate": 0.0001261802575107296, |
| "loss": 0.0479, |
| "mean_token_accuracy": 0.9873545721173287, |
| "num_tokens": 298439.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.8163265306122449, |
| "eval_entropy": 0.0524380611705369, |
| "eval_loss": 0.05050847306847572, |
| "eval_mean_token_accuracy": 0.9866128844776373, |
| "eval_num_tokens": 298439.0, |
| "eval_runtime": 34.3005, |
| "eval_samples_per_second": 10.087, |
| "eval_steps_per_second": 2.536, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.05258838403970003, |
| "epoch": 0.8979591836734694, |
| "grad_norm": 0.07861328125, |
| "learning_rate": 0.00011759656652360516, |
| "loss": 0.048, |
| "mean_token_accuracy": 0.9866469159722329, |
| "num_tokens": 328263.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.05178862875327468, |
| "epoch": 0.9795918367346939, |
| "grad_norm": 0.09912109375, |
| "learning_rate": 0.0001090128755364807, |
| "loss": 0.0476, |
| "mean_token_accuracy": 0.9872021660208702, |
| "num_tokens": 358263.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 0.052568491156163966, |
| "epoch": 1.0571428571428572, |
| "grad_norm": 0.08740234375, |
| "learning_rate": 0.00010042918454935624, |
| "loss": 0.0477, |
| "mean_token_accuracy": 0.987212166974419, |
| "num_tokens": 386372.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 0.050634028390049934, |
| "epoch": 1.1387755102040815, |
| "grad_norm": 0.07861328125, |
| "learning_rate": 9.184549356223176e-05, |
| "loss": 0.0459, |
| "mean_token_accuracy": 0.9879572361707687, |
| "num_tokens": 416328.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 0.04960239427164197, |
| "epoch": 1.220408163265306, |
| "grad_norm": 0.08203125, |
| "learning_rate": 8.32618025751073e-05, |
| "loss": 0.0441, |
| "mean_token_accuracy": 0.9879725813865662, |
| "num_tokens": 446171.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.047286948189139366, |
| "epoch": 1.3020408163265307, |
| "grad_norm": 0.080078125, |
| "learning_rate": 7.467811158798284e-05, |
| "loss": 0.0427, |
| "mean_token_accuracy": 0.9882744088768959, |
| "num_tokens": 476097.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 0.04795173741877079, |
| "epoch": 1.383673469387755, |
| "grad_norm": 0.09033203125, |
| "learning_rate": 6.609442060085838e-05, |
| "loss": 0.043, |
| "mean_token_accuracy": 0.9885840907692909, |
| "num_tokens": 505889.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.04705191999673843, |
| "epoch": 1.4653061224489796, |
| "grad_norm": 0.0791015625, |
| "learning_rate": 5.751072961373391e-05, |
| "loss": 0.0434, |
| "mean_token_accuracy": 0.9881132692098618, |
| "num_tokens": 535679.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.04972094791010022, |
| "epoch": 1.546938775510204, |
| "grad_norm": 0.0966796875, |
| "learning_rate": 4.8927038626609446e-05, |
| "loss": 0.0452, |
| "mean_token_accuracy": 0.9873160928487777, |
| "num_tokens": 565406.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.049302170518785715, |
| "epoch": 1.6285714285714286, |
| "grad_norm": 0.10546875, |
| "learning_rate": 4.034334763948498e-05, |
| "loss": 0.0453, |
| "mean_token_accuracy": 0.9879378706216813, |
| "num_tokens": 595418.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 1.6285714285714286, |
| "eval_entropy": 0.04912949901545185, |
| "eval_loss": 0.04782823845744133, |
| "eval_mean_token_accuracy": 0.9871041678834236, |
| "eval_num_tokens": 595418.0, |
| "eval_runtime": 34.3092, |
| "eval_samples_per_second": 10.085, |
| "eval_steps_per_second": 2.536, |
| "step": 200 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 246, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 2, |
| "save_steps": 200, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.792364691021824e+16, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
