| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.6298815822625347, | |
| "eval_steps": 500, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.000975, | |
| "loss": 1.0238, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_loss": 0.8929525017738342, | |
| "eval_runtime": 4.0089, | |
| "eval_samples_per_second": 74.833, | |
| "eval_steps_per_second": 37.417, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00095, | |
| "loss": 0.9971, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_loss": 0.8510345220565796, | |
| "eval_runtime": 4.0089, | |
| "eval_samples_per_second": 74.833, | |
| "eval_steps_per_second": 37.417, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.000925, | |
| "loss": 0.9522, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 0.8565580248832703, | |
| "eval_runtime": 4.0278, | |
| "eval_samples_per_second": 74.483, | |
| "eval_steps_per_second": 37.241, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0009000000000000001, | |
| "loss": 0.9143, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.8810315132141113, | |
| "eval_runtime": 4.0097, | |
| "eval_samples_per_second": 74.819, | |
| "eval_steps_per_second": 37.409, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.000875, | |
| "loss": 0.9151, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_loss": 0.8461073040962219, | |
| "eval_runtime": 4.0108, | |
| "eval_samples_per_second": 74.798, | |
| "eval_steps_per_second": 37.399, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00085, | |
| "loss": 0.9263, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_loss": 0.8521466255187988, | |
| "eval_runtime": 4.0059, | |
| "eval_samples_per_second": 74.89, | |
| "eval_steps_per_second": 37.445, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.000825, | |
| "loss": 0.9102, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 0.8521981835365295, | |
| "eval_runtime": 4.0413, | |
| "eval_samples_per_second": 74.234, | |
| "eval_steps_per_second": 37.117, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0008, | |
| "loss": 0.8866, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.8259674906730652, | |
| "eval_runtime": 4.0316, | |
| "eval_samples_per_second": 74.412, | |
| "eval_steps_per_second": 37.206, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.0007750000000000001, | |
| "loss": 0.911, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_loss": 0.8008900284767151, | |
| "eval_runtime": 4.0224, | |
| "eval_samples_per_second": 74.583, | |
| "eval_steps_per_second": 37.292, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00075, | |
| "loss": 0.8536, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eval_loss": 0.808663547039032, | |
| "eval_runtime": 5.0615, | |
| "eval_samples_per_second": 59.271, | |
| "eval_steps_per_second": 29.635, | |
| "step": 5000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 20000, | |
| "num_train_epochs": 3, | |
| "save_steps": 5000, | |
| "total_flos": 8851794099093504.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |