| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.861635220125786, | |
| "eval_steps": 500, | |
| "global_step": 2500, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6745161290322581, | |
| "eval_loss": 0.410736620426178, | |
| "eval_runtime": 5.4591, | |
| "eval_samples_per_second": 567.857, | |
| "eval_steps_per_second": 11.907, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.650593990216632e-05, | |
| "loss": 0.6418, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8358064516129032, | |
| "eval_loss": 0.1440049111843109, | |
| "eval_runtime": 5.6568, | |
| "eval_samples_per_second": 548.014, | |
| "eval_steps_per_second": 11.491, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8970967741935484, | |
| "eval_loss": 0.07408486306667328, | |
| "eval_runtime": 5.5564, | |
| "eval_samples_per_second": 557.912, | |
| "eval_steps_per_second": 11.698, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 1.3011879804332637e-05, | |
| "loss": 0.1669, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9135483870967742, | |
| "eval_loss": 0.05311045050621033, | |
| "eval_runtime": 5.5964, | |
| "eval_samples_per_second": 553.925, | |
| "eval_steps_per_second": 11.615, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 9.517819706498952e-06, | |
| "loss": 0.083, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9241935483870968, | |
| "eval_loss": 0.04374333471059799, | |
| "eval_runtime": 5.5916, | |
| "eval_samples_per_second": 554.402, | |
| "eval_steps_per_second": 11.625, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9241935483870968, | |
| "eval_loss": 0.03905611112713814, | |
| "eval_runtime": 5.5141, | |
| "eval_samples_per_second": 562.199, | |
| "eval_steps_per_second": 11.788, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 6.02375960866527e-06, | |
| "loss": 0.0626, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9283870967741935, | |
| "eval_loss": 0.03628788888454437, | |
| "eval_runtime": 5.5615, | |
| "eval_samples_per_second": 557.399, | |
| "eval_steps_per_second": 11.687, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "learning_rate": 2.5296995108315863e-06, | |
| "loss": 0.0545, | |
| "step": 2500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2862, | |
| "num_train_epochs": 9, | |
| "save_steps": 500, | |
| "total_flos": 649636246191744.0, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.5725590574536615, | |
| "num_train_epochs": 9, | |
| "temperature": 2 | |
| } | |
| } | |