| { |
| "best_metric": 0.22694912552833557, |
| "best_model_checkpoint": "./beans_outputs/checkpoint-119", |
| "epoch": 7.0, |
| "eval_steps": 500, |
| "global_step": 119, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.5882352941176471, |
| "grad_norm": 1.3620986938476562, |
| "learning_rate": 1.831932773109244e-05, |
| "loss": 0.9859, |
| "step": 10 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.9323308270676691, |
| "eval_loss": 0.7491994500160217, |
| "eval_runtime": 7.0157, |
| "eval_samples_per_second": 18.958, |
| "eval_steps_per_second": 0.428, |
| "step": 17 |
| }, |
| { |
| "epoch": 1.1764705882352942, |
| "grad_norm": 1.1871144771575928, |
| "learning_rate": 1.6638655462184877e-05, |
| "loss": 0.8078, |
| "step": 20 |
| }, |
| { |
| "epoch": 1.7647058823529411, |
| "grad_norm": 1.1515430212020874, |
| "learning_rate": 1.4957983193277313e-05, |
| "loss": 0.6759, |
| "step": 30 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.9624060150375939, |
| "eval_loss": 0.5281645059585571, |
| "eval_runtime": 7.0064, |
| "eval_samples_per_second": 18.983, |
| "eval_steps_per_second": 0.428, |
| "step": 34 |
| }, |
| { |
| "epoch": 2.3529411764705883, |
| "grad_norm": 1.2202690839767456, |
| "learning_rate": 1.3277310924369749e-05, |
| "loss": 0.5495, |
| "step": 40 |
| }, |
| { |
| "epoch": 2.9411764705882355, |
| "grad_norm": 1.0186351537704468, |
| "learning_rate": 1.1596638655462186e-05, |
| "loss": 0.461, |
| "step": 50 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.9624060150375939, |
| "eval_loss": 0.37340047955513, |
| "eval_runtime": 7.3765, |
| "eval_samples_per_second": 18.03, |
| "eval_steps_per_second": 0.407, |
| "step": 51 |
| }, |
| { |
| "epoch": 3.5294117647058822, |
| "grad_norm": 1.1645742654800415, |
| "learning_rate": 9.915966386554622e-06, |
| "loss": 0.4048, |
| "step": 60 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.9699248120300752, |
| "eval_loss": 0.2986946403980255, |
| "eval_runtime": 7.0493, |
| "eval_samples_per_second": 18.867, |
| "eval_steps_per_second": 0.426, |
| "step": 68 |
| }, |
| { |
| "epoch": 4.117647058823529, |
| "grad_norm": 1.0221604108810425, |
| "learning_rate": 8.23529411764706e-06, |
| "loss": 0.3271, |
| "step": 70 |
| }, |
| { |
| "epoch": 4.705882352941177, |
| "grad_norm": 1.340811014175415, |
| "learning_rate": 6.5546218487394966e-06, |
| "loss": 0.3174, |
| "step": 80 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.9699248120300752, |
| "eval_loss": 0.25426048040390015, |
| "eval_runtime": 7.1157, |
| "eval_samples_per_second": 18.691, |
| "eval_steps_per_second": 0.422, |
| "step": 85 |
| }, |
| { |
| "epoch": 5.294117647058823, |
| "grad_norm": 1.0393356084823608, |
| "learning_rate": 4.873949579831933e-06, |
| "loss": 0.2772, |
| "step": 90 |
| }, |
| { |
| "epoch": 5.882352941176471, |
| "grad_norm": 0.8360141515731812, |
| "learning_rate": 3.1932773109243696e-06, |
| "loss": 0.2542, |
| "step": 100 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.9774436090225563, |
| "eval_loss": 0.22818955779075623, |
| "eval_runtime": 7.1818, |
| "eval_samples_per_second": 18.519, |
| "eval_steps_per_second": 0.418, |
| "step": 102 |
| }, |
| { |
| "epoch": 6.470588235294118, |
| "grad_norm": 1.0879504680633545, |
| "learning_rate": 1.5126050420168068e-06, |
| "loss": 0.2627, |
| "step": 110 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.9699248120300752, |
| "eval_loss": 0.22694912552833557, |
| "eval_runtime": 7.2426, |
| "eval_samples_per_second": 18.364, |
| "eval_steps_per_second": 0.414, |
| "step": 119 |
| }, |
| { |
| "epoch": 7.0, |
| "step": 119, |
| "total_flos": 5.608920478833746e+17, |
| "train_loss": 0.46666095236770244, |
| "train_runtime": 510.5808, |
| "train_samples_per_second": 14.176, |
| "train_steps_per_second": 0.233 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 119, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 7, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 5.608920478833746e+17, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
