| { | |
| "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": 0.6954838709677419, | |
| "eval_loss": 0.4076870381832123, | |
| "eval_runtime": 5.8135, | |
| "eval_samples_per_second": 533.239, | |
| "eval_steps_per_second": 11.181, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.650593990216632e-05, | |
| "loss": 0.6428, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8416129032258064, | |
| "eval_loss": 0.13920199871063232, | |
| "eval_runtime": 5.9217, | |
| "eval_samples_per_second": 523.497, | |
| "eval_steps_per_second": 10.977, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8987096774193548, | |
| "eval_loss": 0.07036732137203217, | |
| "eval_runtime": 5.6928, | |
| "eval_samples_per_second": 544.549, | |
| "eval_steps_per_second": 11.418, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 1.3011879804332637e-05, | |
| "loss": 0.163, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9196774193548387, | |
| "eval_loss": 0.05059635266661644, | |
| "eval_runtime": 5.7158, | |
| "eval_samples_per_second": 542.352, | |
| "eval_steps_per_second": 11.372, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 9.517819706498952e-06, | |
| "loss": 0.0813, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9267741935483871, | |
| "eval_loss": 0.04242779687047005, | |
| "eval_runtime": 5.8907, | |
| "eval_samples_per_second": 526.252, | |
| "eval_steps_per_second": 11.034, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9280645161290323, | |
| "eval_loss": 0.03737134486436844, | |
| "eval_runtime": 5.7113, | |
| "eval_samples_per_second": 542.78, | |
| "eval_steps_per_second": 11.381, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 6.02375960866527e-06, | |
| "loss": 0.0615, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9316129032258065, | |
| "eval_loss": 0.035418927669525146, | |
| "eval_runtime": 5.7762, | |
| "eval_samples_per_second": 536.689, | |
| "eval_steps_per_second": 11.253, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "learning_rate": 2.5296995108315863e-06, | |
| "loss": 0.0536, | |
| "step": 2500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2862, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 500, | |
| "total_flos": 649131947564688.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.9937737785515707, | |
| "num_train_epochs": 9, | |
| "temperature": 2 | |
| } | |
| } | |