| { |
| "best_global_step": 100, |
| "best_metric": 0.11088702827692032, |
| "best_model_checkpoint": "/content/models/gemma_qlora_lmh_inst/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": 2.139886352419853, |
| "epoch": 0.3418803418803419, |
| "grad_norm": 6.644908428192139, |
| "learning_rate": 8.389830508474577e-06, |
| "loss": 0.449, |
| "mean_token_accuracy": 0.9069619461894035, |
| "num_tokens": 118924.0, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.3418803418803419, |
| "eval_entropy": 2.112155056407309, |
| "eval_loss": 0.16615918278694153, |
| "eval_mean_token_accuracy": 0.9459472375038342, |
| "eval_num_tokens": 118924.0, |
| "eval_runtime": 47.2708, |
| "eval_samples_per_second": 39.369, |
| "eval_steps_per_second": 2.475, |
| "step": 20 |
| }, |
| { |
| "entropy": 2.0837366968393325, |
| "epoch": 0.6837606837606838, |
| "grad_norm": 2.919997215270996, |
| "learning_rate": 6.694915254237288e-06, |
| "loss": 0.1523, |
| "mean_token_accuracy": 0.9496713325381279, |
| "num_tokens": 236856.0, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.6837606837606838, |
| "eval_entropy": 2.0886894973934207, |
| "eval_loss": 0.1265888810157776, |
| "eval_mean_token_accuracy": 0.959908177709987, |
| "eval_num_tokens": 236856.0, |
| "eval_runtime": 46.4967, |
| "eval_samples_per_second": 40.024, |
| "eval_steps_per_second": 2.516, |
| "step": 40 |
| }, |
| { |
| "entropy": 2.087970663339664, |
| "epoch": 1.017094017094017, |
| "grad_norm": 2.6256933212280273, |
| "learning_rate": 5e-06, |
| "loss": 0.136, |
| "mean_token_accuracy": 0.9595257181387681, |
| "num_tokens": 347411.0, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.017094017094017, |
| "eval_entropy": 2.1006248659557767, |
| "eval_loss": 0.12076133489608765, |
| "eval_mean_token_accuracy": 0.960268861717648, |
| "eval_num_tokens": 347411.0, |
| "eval_runtime": 46.4488, |
| "eval_samples_per_second": 40.066, |
| "eval_steps_per_second": 2.519, |
| "step": 60 |
| }, |
| { |
| "entropy": 2.077034068107605, |
| "epoch": 1.358974358974359, |
| "grad_norm": 2.3626747131347656, |
| "learning_rate": 3.305084745762712e-06, |
| "loss": 0.096, |
| "mean_token_accuracy": 0.9696433156728744, |
| "num_tokens": 463138.0, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.358974358974359, |
| "eval_entropy": 2.0711163577870426, |
| "eval_loss": 0.12328439950942993, |
| "eval_mean_token_accuracy": 0.9593201702476567, |
| "eval_num_tokens": 463138.0, |
| "eval_runtime": 46.4721, |
| "eval_samples_per_second": 40.046, |
| "eval_steps_per_second": 2.518, |
| "step": 80 |
| }, |
| { |
| "entropy": 2.0657004177570344, |
| "epoch": 1.7008547008547008, |
| "grad_norm": 1.6588151454925537, |
| "learning_rate": 1.6101694915254237e-06, |
| "loss": 0.0803, |
| "mean_token_accuracy": 0.9724445611238479, |
| "num_tokens": 581349.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.7008547008547008, |
| "eval_entropy": 2.0421143230210004, |
| "eval_loss": 0.11088702827692032, |
| "eval_mean_token_accuracy": 0.9651837888945881, |
| "eval_num_tokens": 581349.0, |
| "eval_runtime": 46.4848, |
| "eval_samples_per_second": 40.035, |
| "eval_steps_per_second": 2.517, |
| "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": 1.6519515681526272e+16, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
