| { | |
| "best_metric": 2.2463176250457764, | |
| "best_model_checkpoint": "./outputs/checkpoint-1600", | |
| "epoch": 1.1657559198542806, | |
| "eval_steps": 100, | |
| "global_step": 1600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0002, | |
| "loss": 2.7205, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 2.6170873641967773, | |
| "eval_runtime": 228.505, | |
| "eval_samples_per_second": 27.457, | |
| "eval_steps_per_second": 3.435, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5807, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 2.5663836002349854, | |
| "eval_runtime": 210.5353, | |
| "eval_samples_per_second": 29.8, | |
| "eval_steps_per_second": 3.729, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5398, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 2.531459331512451, | |
| "eval_runtime": 245.5646, | |
| "eval_samples_per_second": 25.549, | |
| "eval_steps_per_second": 3.197, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5133, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 2.497565746307373, | |
| "eval_runtime": 244.6469, | |
| "eval_samples_per_second": 25.645, | |
| "eval_steps_per_second": 3.209, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4697, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 2.4726672172546387, | |
| "eval_runtime": 244.9375, | |
| "eval_samples_per_second": 25.615, | |
| "eval_steps_per_second": 3.205, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4435, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 2.4451191425323486, | |
| "eval_runtime": 216.439, | |
| "eval_samples_per_second": 28.987, | |
| "eval_steps_per_second": 3.627, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002, | |
| "loss": 2.426, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 2.4199516773223877, | |
| "eval_runtime": 206.0859, | |
| "eval_samples_per_second": 30.444, | |
| "eval_steps_per_second": 3.809, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4041, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 2.399696111679077, | |
| "eval_runtime": 205.7858, | |
| "eval_samples_per_second": 30.488, | |
| "eval_steps_per_second": 3.815, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0002, | |
| "loss": 2.369, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 2.380272150039673, | |
| "eval_runtime": 206.0924, | |
| "eval_samples_per_second": 30.443, | |
| "eval_steps_per_second": 3.809, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3663, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 2.361511707305908, | |
| "eval_runtime": 206.0953, | |
| "eval_samples_per_second": 30.442, | |
| "eval_steps_per_second": 3.809, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3596, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 2.337871551513672, | |
| "eval_runtime": 206.2542, | |
| "eval_samples_per_second": 30.419, | |
| "eval_steps_per_second": 3.806, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3156, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 2.318875789642334, | |
| "eval_runtime": 205.5733, | |
| "eval_samples_per_second": 30.52, | |
| "eval_steps_per_second": 3.819, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3267, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 2.3044281005859375, | |
| "eval_runtime": 206.2612, | |
| "eval_samples_per_second": 30.418, | |
| "eval_steps_per_second": 3.806, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2788, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 2.2812631130218506, | |
| "eval_runtime": 205.7511, | |
| "eval_samples_per_second": 30.493, | |
| "eval_steps_per_second": 3.815, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2397, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_loss": 2.2644264698028564, | |
| "eval_runtime": 206.0922, | |
| "eval_samples_per_second": 30.443, | |
| "eval_steps_per_second": 3.809, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.0002, | |
| "loss": 2.232, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_loss": 2.2463176250457764, | |
| "eval_runtime": 205.9251, | |
| "eval_samples_per_second": 30.467, | |
| "eval_steps_per_second": 3.812, | |
| "step": 1600 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4116, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 4.70673370653696e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |