| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "global_step": 19200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.535132032339264e-05, | |
| "loss": 2.3386, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7039312338205812, | |
| "eval_loss": 1.4716728925704956, | |
| "eval_runtime": 48.9597, | |
| "eval_samples_per_second": 155.311, | |
| "eval_steps_per_second": 0.98, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.6900880215595094e-05, | |
| "loss": 1.276, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7318570778556786, | |
| "eval_loss": 1.2896931171417236, | |
| "eval_runtime": 45.4592, | |
| "eval_samples_per_second": 167.271, | |
| "eval_steps_per_second": 1.056, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 1.7807314645155048e-05, | |
| "loss": 1.1451, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7376776063935869, | |
| "eval_loss": 1.2503294944763184, | |
| "eval_runtime": 45.5404, | |
| "eval_samples_per_second": 166.973, | |
| "eval_steps_per_second": 1.054, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 1.8450440107797548e-05, | |
| "loss": 1.0936, | |
| "step": 3840 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7428573517905067, | |
| "eval_loss": 1.2241426706314087, | |
| "eval_runtime": 45.7082, | |
| "eval_samples_per_second": 166.36, | |
| "eval_steps_per_second": 1.05, | |
| "step": 3840 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 1.894928697180815e-05, | |
| "loss": 1.0595, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7445576106276123, | |
| "eval_loss": 1.203609824180603, | |
| "eval_runtime": 45.413, | |
| "eval_samples_per_second": 167.441, | |
| "eval_steps_per_second": 1.057, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 1.93568745373575e-05, | |
| "loss": 1.0345, | |
| "step": 5760 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7489516952862771, | |
| "eval_loss": 1.181204080581665, | |
| "eval_runtime": 45.4395, | |
| "eval_samples_per_second": 167.343, | |
| "eval_steps_per_second": 1.056, | |
| "step": 5760 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 1.9701484913790247e-05, | |
| "loss": 1.0157, | |
| "step": 6720 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7509223157745868, | |
| "eval_loss": 1.1679136753082275, | |
| "eval_runtime": 45.4562, | |
| "eval_samples_per_second": 167.282, | |
| "eval_steps_per_second": 1.056, | |
| "step": 6720 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9947, | |
| "step": 7680 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7525814641372622, | |
| "eval_loss": 1.1508731842041016, | |
| "eval_runtime": 45.8039, | |
| "eval_samples_per_second": 166.012, | |
| "eval_steps_per_second": 1.048, | |
| "step": 7680 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9781, | |
| "step": 8640 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.751431242250465, | |
| "eval_loss": 1.155458688735962, | |
| "eval_runtime": 45.6044, | |
| "eval_samples_per_second": 166.738, | |
| "eval_steps_per_second": 1.053, | |
| "step": 8640 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9671, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7548351121589497, | |
| "eval_loss": 1.1350805759429932, | |
| "eval_runtime": 45.4513, | |
| "eval_samples_per_second": 167.3, | |
| "eval_steps_per_second": 1.056, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9568, | |
| "step": 10560 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.7557078568184633, | |
| "eval_loss": 1.132196068763733, | |
| "eval_runtime": 45.4933, | |
| "eval_samples_per_second": 167.146, | |
| "eval_steps_per_second": 1.055, | |
| "step": 10560 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9456, | |
| "step": 11520 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.757524201934202, | |
| "eval_loss": 1.1184238195419312, | |
| "eval_runtime": 46.0479, | |
| "eval_samples_per_second": 165.132, | |
| "eval_steps_per_second": 1.042, | |
| "step": 11520 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9372, | |
| "step": 12480 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.7570365161728335, | |
| "eval_loss": 1.1217310428619385, | |
| "eval_runtime": 45.5154, | |
| "eval_samples_per_second": 167.064, | |
| "eval_steps_per_second": 1.055, | |
| "step": 12480 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9269, | |
| "step": 13440 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.7585645249647832, | |
| "eval_loss": 1.1052144765853882, | |
| "eval_runtime": 45.5094, | |
| "eval_samples_per_second": 167.086, | |
| "eval_steps_per_second": 1.055, | |
| "step": 13440 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9181, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.7602352596641937, | |
| "eval_loss": 1.10312819480896, | |
| "eval_runtime": 45.4675, | |
| "eval_samples_per_second": 167.24, | |
| "eval_steps_per_second": 1.056, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9113, | |
| "step": 15360 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7619846227187718, | |
| "eval_loss": 1.0922868251800537, | |
| "eval_runtime": 46.139, | |
| "eval_samples_per_second": 164.806, | |
| "eval_steps_per_second": 1.04, | |
| "step": 15360 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.903, | |
| "step": 16320 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.7612785401524097, | |
| "eval_loss": 1.097521424293518, | |
| "eval_runtime": 45.4875, | |
| "eval_samples_per_second": 167.167, | |
| "eval_steps_per_second": 1.055, | |
| "step": 16320 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.897, | |
| "step": 17280 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.7631071933982371, | |
| "eval_loss": 1.084978699684143, | |
| "eval_runtime": 45.4666, | |
| "eval_samples_per_second": 167.244, | |
| "eval_steps_per_second": 1.056, | |
| "step": 17280 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8898, | |
| "step": 18240 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.7632023931815797, | |
| "eval_loss": 1.0815385580062866, | |
| "eval_runtime": 45.8568, | |
| "eval_samples_per_second": 165.821, | |
| "eval_steps_per_second": 1.047, | |
| "step": 18240 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8846, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7635695694279448, | |
| "eval_loss": 1.0812164545059204, | |
| "eval_runtime": 46.0156, | |
| "eval_samples_per_second": 165.248, | |
| "eval_steps_per_second": 1.043, | |
| "step": 19200 | |
| } | |
| ], | |
| "max_steps": 38400, | |
| "num_train_epochs": 40, | |
| "total_flos": 3062980304437248.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |