| { |
| "best_global_step": 100, |
| "best_metric": 0.12709300220012665, |
| "best_model_checkpoint": "/content/models/qwen_qlora_lmh/checkpoint-100", |
| "epoch": 1.7008547008547008, |
| "eval_steps": 20, |
| "global_step": 100, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 0.5939132001250982, |
| "epoch": 0.3418803418803419, |
| "grad_norm": 3.109375, |
| "learning_rate": 8.389830508474577e-06, |
| "loss": 0.4929, |
| "mean_token_accuracy": 0.8503338515758514, |
| "num_tokens": 107898.0, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.3418803418803419, |
| "eval_entropy": 0.18380238570399204, |
| "eval_loss": 0.18679018318653107, |
| "eval_mean_token_accuracy": 0.945185255701259, |
| "eval_num_tokens": 107898.0, |
| "eval_runtime": 21.0263, |
| "eval_samples_per_second": 89.079, |
| "eval_steps_per_second": 5.612, |
| "step": 20 |
| }, |
| { |
| "entropy": 0.13663306124508381, |
| "epoch": 0.6837606837606838, |
| "grad_norm": 1.28125, |
| "learning_rate": 6.694915254237288e-06, |
| "loss": 0.1401, |
| "mean_token_accuracy": 0.9570734396576881, |
| "num_tokens": 215825.0, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.6837606837606838, |
| "eval_entropy": 0.1707214197870028, |
| "eval_loss": 0.14553521573543549, |
| "eval_mean_token_accuracy": 0.9526250013860605, |
| "eval_num_tokens": 215825.0, |
| "eval_runtime": 21.0083, |
| "eval_samples_per_second": 89.155, |
| "eval_steps_per_second": 5.617, |
| "step": 40 |
| }, |
| { |
| "entropy": 0.16002595195403466, |
| "epoch": 1.017094017094017, |
| "grad_norm": 1.6640625, |
| "learning_rate": 5e-06, |
| "loss": 0.1523, |
| "mean_token_accuracy": 0.950343770858569, |
| "num_tokens": 324179.0, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.017094017094017, |
| "eval_entropy": 0.1480934122987723, |
| "eval_loss": 0.1345394253730774, |
| "eval_mean_token_accuracy": 0.9556088796106436, |
| "eval_num_tokens": 324179.0, |
| "eval_runtime": 21.4759, |
| "eval_samples_per_second": 87.214, |
| "eval_steps_per_second": 5.495, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.12568409610539674, |
| "epoch": 1.358974358974359, |
| "grad_norm": 2.6875, |
| "learning_rate": 3.305084745762712e-06, |
| "loss": 0.1282, |
| "mean_token_accuracy": 0.9555983245372772, |
| "num_tokens": 432702.0, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.358974358974359, |
| "eval_entropy": 0.11805923264157975, |
| "eval_loss": 0.12935617566108704, |
| "eval_mean_token_accuracy": 0.9543270655607773, |
| "eval_num_tokens": 432702.0, |
| "eval_runtime": 21.0319, |
| "eval_samples_per_second": 89.055, |
| "eval_steps_per_second": 5.611, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.11634381394833326, |
| "epoch": 1.7008547008547008, |
| "grad_norm": 2.6875, |
| "learning_rate": 1.6101694915254237e-06, |
| "loss": 0.1122, |
| "mean_token_accuracy": 0.9652405023574829, |
| "num_tokens": 542919.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.7008547008547008, |
| "eval_entropy": 0.11054411360015304, |
| "eval_loss": 0.12709300220012665, |
| "eval_mean_token_accuracy": 0.9555371065261, |
| "eval_num_tokens": 542919.0, |
| "eval_runtime": 21.037, |
| "eval_samples_per_second": 89.034, |
| "eval_steps_per_second": 5.609, |
| "step": 100 |
| } |
| ], |
| "logging_steps": 20, |
| "max_steps": 118, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 2, |
| "save_steps": 20, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.1244141143660544e+16, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
