| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.289308176100629, | |
| "eval_steps": 500, | |
| "global_step": 2000, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5838709677419355, | |
| "eval_loss": 0.19487537443637848, | |
| "eval_runtime": 1.4529, | |
| "eval_samples_per_second": 2133.723, | |
| "eval_steps_per_second": 44.739, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.650593990216632e-05, | |
| "loss": 0.3137, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8280645161290323, | |
| "eval_loss": 0.09586000442504883, | |
| "eval_runtime": 1.4395, | |
| "eval_samples_per_second": 2153.495, | |
| "eval_steps_per_second": 45.154, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8774193548387097, | |
| "eval_loss": 0.06496233493089676, | |
| "eval_runtime": 1.4506, | |
| "eval_samples_per_second": 2136.985, | |
| "eval_steps_per_second": 44.808, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 1.3011879804332637e-05, | |
| "loss": 0.1102, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8958064516129032, | |
| "eval_loss": 0.04958318546414375, | |
| "eval_runtime": 1.4538, | |
| "eval_samples_per_second": 2132.391, | |
| "eval_steps_per_second": 44.711, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 9.517819706498952e-06, | |
| "loss": 0.0707, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9080645161290323, | |
| "eval_loss": 0.04088330268859863, | |
| "eval_runtime": 1.4623, | |
| "eval_samples_per_second": 2119.972, | |
| "eval_steps_per_second": 44.451, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9138709677419354, | |
| "eval_loss": 0.03609682619571686, | |
| "eval_runtime": 1.4689, | |
| "eval_samples_per_second": 2110.457, | |
| "eval_steps_per_second": 44.252, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 6.02375960866527e-06, | |
| "loss": 0.0561, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2862, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 500, | |
| "total_flos": 603592017533988.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.08306622214416404, | |
| "num_train_epochs": 9, | |
| "temperature": 14 | |
| } | |
| } | |