| { |
| "best_global_step": 100, |
| "best_metric": 0.28937000036239624, |
| "best_model_checkpoint": "/content/models/gemma_qlora_lmh_inst/checkpoint-100", |
| "epoch": 1.9607843137254903, |
| "eval_steps": 20, |
| "global_step": 100, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 2.1317927479743957, |
| "epoch": 0.39215686274509803, |
| "grad_norm": 4.763621807098389, |
| "learning_rate": 8.137254901960784e-06, |
| "loss": 0.7536, |
| "mean_token_accuracy": 0.76171875, |
| "num_tokens": 46761.0, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.39215686274509803, |
| "eval_entropy": 2.128274752543523, |
| "eval_loss": 0.34577950835227966, |
| "eval_mean_token_accuracy": 0.7480332163664011, |
| "eval_num_tokens": 46761.0, |
| "eval_runtime": 2.4531, |
| "eval_samples_per_second": 82.754, |
| "eval_steps_per_second": 5.3, |
| "step": 20 |
| }, |
| { |
| "entropy": 2.1670219361782075, |
| "epoch": 0.7843137254901961, |
| "grad_norm": 5.078281402587891, |
| "learning_rate": 6.176470588235295e-06, |
| "loss": 0.3072, |
| "mean_token_accuracy": 0.8203125, |
| "num_tokens": 95614.0, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.7843137254901961, |
| "eval_entropy": 2.1048440016233005, |
| "eval_loss": 0.35016557574272156, |
| "eval_mean_token_accuracy": 0.8236451057287363, |
| "eval_num_tokens": 95614.0, |
| "eval_runtime": 2.4636, |
| "eval_samples_per_second": 82.4, |
| "eval_steps_per_second": 5.277, |
| "step": 40 |
| }, |
| { |
| "entropy": 2.1607252150774, |
| "epoch": 1.1764705882352942, |
| "grad_norm": 10.205085754394531, |
| "learning_rate": 4.215686274509805e-06, |
| "loss": 0.266, |
| "mean_token_accuracy": 0.86328125, |
| "num_tokens": 143455.0, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.1764705882352942, |
| "eval_entropy": 2.12537236397083, |
| "eval_loss": 0.3281736969947815, |
| "eval_mean_token_accuracy": 0.835664336497967, |
| "eval_num_tokens": 143455.0, |
| "eval_runtime": 2.4452, |
| "eval_samples_per_second": 83.019, |
| "eval_steps_per_second": 5.317, |
| "step": 60 |
| }, |
| { |
| "entropy": 2.1871365696191787, |
| "epoch": 1.5686274509803921, |
| "grad_norm": 6.319624423980713, |
| "learning_rate": 2.254901960784314e-06, |
| "loss": 0.2363, |
| "mean_token_accuracy": 0.88125, |
| "num_tokens": 192572.0, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.5686274509803921, |
| "eval_entropy": 2.1334355702767005, |
| "eval_loss": 0.29352903366088867, |
| "eval_mean_token_accuracy": 0.8465909086740934, |
| "eval_num_tokens": 192572.0, |
| "eval_runtime": 2.4718, |
| "eval_samples_per_second": 82.128, |
| "eval_steps_per_second": 5.259, |
| "step": 80 |
| }, |
| { |
| "entropy": 2.1668645977973937, |
| "epoch": 1.9607843137254903, |
| "grad_norm": 3.9370977878570557, |
| "learning_rate": 2.9411764705882356e-07, |
| "loss": 0.2172, |
| "mean_token_accuracy": 0.8984375, |
| "num_tokens": 240555.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.9607843137254903, |
| "eval_entropy": 2.129720770395719, |
| "eval_loss": 0.28937000036239624, |
| "eval_mean_token_accuracy": 0.8597027980364286, |
| "eval_num_tokens": 240555.0, |
| "eval_runtime": 2.4499, |
| "eval_samples_per_second": 82.861, |
| "eval_steps_per_second": 5.306, |
| "step": 100 |
| } |
| ], |
| "logging_steps": 20, |
| "max_steps": 102, |
| "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": 6060538554167808.0, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
