| { |
| "best_global_step": 100, |
| "best_metric": 0.268216073513031, |
| "best_model_checkpoint": "/content/models/gemma_qlora_lmh/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": 2.652500593662262, |
| "epoch": 0.39215686274509803, |
| "grad_norm": 28.916894912719727, |
| "learning_rate": 8.137254901960784e-06, |
| "loss": 0.68, |
| "mean_token_accuracy": 0.71015625, |
| "num_tokens": 48092.0, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.39215686274509803, |
| "eval_entropy": 2.5002221510960507, |
| "eval_loss": 0.31822389364242554, |
| "eval_mean_token_accuracy": 0.8417832163664011, |
| "eval_num_tokens": 48092.0, |
| "eval_runtime": 2.4841, |
| "eval_samples_per_second": 81.719, |
| "eval_steps_per_second": 5.233, |
| "step": 20 |
| }, |
| { |
| "entropy": 2.523895502090454, |
| "epoch": 0.7843137254901961, |
| "grad_norm": 2.8356525897979736, |
| "learning_rate": 6.176470588235295e-06, |
| "loss": 0.3528, |
| "mean_token_accuracy": 0.79765625, |
| "num_tokens": 98284.0, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.7843137254901961, |
| "eval_entropy": 2.480310696821946, |
| "eval_loss": 0.3085189759731293, |
| "eval_mean_token_accuracy": 0.831075173157912, |
| "eval_num_tokens": 98284.0, |
| "eval_runtime": 2.4982, |
| "eval_samples_per_second": 81.258, |
| "eval_steps_per_second": 5.204, |
| "step": 40 |
| }, |
| { |
| "entropy": 2.519091600179672, |
| "epoch": 1.1764705882352942, |
| "grad_norm": 8.396736145019531, |
| "learning_rate": 4.215686274509805e-06, |
| "loss": 0.2831, |
| "mean_token_accuracy": 0.8506696432828903, |
| "num_tokens": 147466.0, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.1764705882352942, |
| "eval_entropy": 2.4424293774824877, |
| "eval_loss": 0.2718758285045624, |
| "eval_mean_token_accuracy": 0.8610139855971704, |
| "eval_num_tokens": 147466.0, |
| "eval_runtime": 2.5004, |
| "eval_samples_per_second": 81.187, |
| "eval_steps_per_second": 5.199, |
| "step": 60 |
| }, |
| { |
| "entropy": 2.479978394508362, |
| "epoch": 1.5686274509803921, |
| "grad_norm": 9.535552978515625, |
| "learning_rate": 2.254901960784314e-06, |
| "loss": 0.2497, |
| "mean_token_accuracy": 0.8796875, |
| "num_tokens": 197917.0, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.5686274509803921, |
| "eval_entropy": 2.43264233149015, |
| "eval_loss": 0.2887224853038788, |
| "eval_mean_token_accuracy": 0.8634178317510165, |
| "eval_num_tokens": 197917.0, |
| "eval_runtime": 2.5098, |
| "eval_samples_per_second": 80.883, |
| "eval_steps_per_second": 5.18, |
| "step": 80 |
| }, |
| { |
| "entropy": 2.4772088170051574, |
| "epoch": 1.9607843137254903, |
| "grad_norm": 3.218280076980591, |
| "learning_rate": 2.9411764705882356e-07, |
| "loss": 0.2359, |
| "mean_token_accuracy": 0.89140625, |
| "num_tokens": 247245.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.9607843137254903, |
| "eval_entropy": 2.4411074198209324, |
| "eval_loss": 0.268216073513031, |
| "eval_mean_token_accuracy": 0.8658216779048626, |
| "eval_num_tokens": 247245.0, |
| "eval_runtime": 2.5076, |
| "eval_samples_per_second": 80.953, |
| "eval_steps_per_second": 5.184, |
| "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": 6148418735678976.0, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
