| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "global_step": 5600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9008928537368774, | |
| "eval_loss": 0.2425430864095688, | |
| "eval_runtime": 15.3273, | |
| "eval_samples_per_second": 73.072, | |
| "eval_steps_per_second": 9.134, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 3.0769035067318356e-05, | |
| "loss": 0.2497, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.90625, | |
| "eval_loss": 0.2671281397342682, | |
| "eval_runtime": 15.4851, | |
| "eval_samples_per_second": 72.327, | |
| "eval_steps_per_second": 9.041, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9285714030265808, | |
| "eval_loss": 0.37120091915130615, | |
| "eval_runtime": 15.3992, | |
| "eval_samples_per_second": 72.731, | |
| "eval_steps_per_second": 9.091, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 2.775246300189499e-05, | |
| "loss": 0.0886, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9357143044471741, | |
| "eval_loss": 0.3282666504383087, | |
| "eval_runtime": 15.4636, | |
| "eval_samples_per_second": 72.428, | |
| "eval_steps_per_second": 9.054, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9303571581840515, | |
| "eval_loss": 0.36973103880882263, | |
| "eval_runtime": 15.65, | |
| "eval_samples_per_second": 71.565, | |
| "eval_steps_per_second": 8.946, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "learning_rate": 2.473589093647162e-05, | |
| "loss": 0.0277, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9303571581840515, | |
| "eval_loss": 0.4028499722480774, | |
| "eval_runtime": 15.6492, | |
| "eval_samples_per_second": 71.569, | |
| "eval_steps_per_second": 8.946, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9285714030265808, | |
| "eval_loss": 0.4771631956100464, | |
| "eval_runtime": 15.5073, | |
| "eval_samples_per_second": 72.224, | |
| "eval_steps_per_second": 9.028, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 2.1719318871048256e-05, | |
| "loss": 0.023, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9205357432365417, | |
| "eval_loss": 0.5438315868377686, | |
| "eval_runtime": 15.6912, | |
| "eval_samples_per_second": 71.378, | |
| "eval_steps_per_second": 8.922, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "learning_rate": 1.8702746805624887e-05, | |
| "loss": 0.0229, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9276785850524902, | |
| "eval_loss": 0.51451176404953, | |
| "eval_runtime": 15.4278, | |
| "eval_samples_per_second": 72.596, | |
| "eval_steps_per_second": 9.075, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9223214387893677, | |
| "eval_loss": 0.4909174144268036, | |
| "eval_runtime": 15.4206, | |
| "eval_samples_per_second": 72.63, | |
| "eval_steps_per_second": 9.079, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 10.71, | |
| "learning_rate": 1.5686174740201518e-05, | |
| "loss": 0.0151, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9205357432365417, | |
| "eval_loss": 0.5145156979560852, | |
| "eval_runtime": 15.4916, | |
| "eval_samples_per_second": 72.297, | |
| "eval_steps_per_second": 9.037, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9241071343421936, | |
| "eval_loss": 0.5320212244987488, | |
| "eval_runtime": 15.5222, | |
| "eval_samples_per_second": 72.155, | |
| "eval_steps_per_second": 9.019, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "learning_rate": 1.2669602674778149e-05, | |
| "loss": 0.0164, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.925000011920929, | |
| "eval_loss": 0.4845915138721466, | |
| "eval_runtime": 15.5092, | |
| "eval_samples_per_second": 72.215, | |
| "eval_steps_per_second": 9.027, | |
| "step": 3640 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9232142567634583, | |
| "eval_loss": 0.5342178344726562, | |
| "eval_runtime": 15.4696, | |
| "eval_samples_per_second": 72.4, | |
| "eval_steps_per_second": 9.05, | |
| "step": 3920 | |
| }, | |
| { | |
| "epoch": 14.29, | |
| "learning_rate": 9.653030609354778e-06, | |
| "loss": 0.0114, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9241071343421936, | |
| "eval_loss": 0.5672621726989746, | |
| "eval_runtime": 15.5099, | |
| "eval_samples_per_second": 72.212, | |
| "eval_steps_per_second": 9.027, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9258928298950195, | |
| "eval_loss": 0.562305748462677, | |
| "eval_runtime": 15.6284, | |
| "eval_samples_per_second": 71.664, | |
| "eval_steps_per_second": 8.958, | |
| "step": 4480 | |
| }, | |
| { | |
| "epoch": 16.07, | |
| "learning_rate": 6.636458543931411e-06, | |
| "loss": 0.0087, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.9258928298950195, | |
| "eval_loss": 0.5690274238586426, | |
| "eval_runtime": 15.6003, | |
| "eval_samples_per_second": 71.793, | |
| "eval_steps_per_second": 8.974, | |
| "step": 4760 | |
| }, | |
| { | |
| "epoch": 17.86, | |
| "learning_rate": 3.619886478508042e-06, | |
| "loss": 0.0086, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.925000011920929, | |
| "eval_loss": 0.5821966528892517, | |
| "eval_runtime": 15.4277, | |
| "eval_samples_per_second": 72.597, | |
| "eval_steps_per_second": 9.075, | |
| "step": 5040 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.925000011920929, | |
| "eval_loss": 0.5867368578910828, | |
| "eval_runtime": 15.4548, | |
| "eval_samples_per_second": 72.469, | |
| "eval_steps_per_second": 9.059, | |
| "step": 5320 | |
| }, | |
| { | |
| "epoch": 19.64, | |
| "learning_rate": 6.033144130846736e-07, | |
| "loss": 0.0067, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9258928298950195, | |
| "eval_loss": 0.5937979221343994, | |
| "eval_runtime": 15.5033, | |
| "eval_samples_per_second": 72.243, | |
| "eval_steps_per_second": 9.03, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 5600, | |
| "total_flos": 2.08752625385472e+16, | |
| "train_loss": 0.042873825313789506, | |
| "train_runtime": 6011.0282, | |
| "train_samples_per_second": 14.906, | |
| "train_steps_per_second": 0.932 | |
| } | |
| ], | |
| "max_steps": 5600, | |
| "num_train_epochs": 20, | |
| "total_flos": 2.08752625385472e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |