| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 8.0, | |
| "global_step": 12336, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 1.9189364461738006e-05, | |
| "loss": 2.5752, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1.8378728923476004e-05, | |
| "loss": 2.4758, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.756809338521401e-05, | |
| "loss": 2.4695, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.0689399242401123, | |
| "eval_runtime": 81.3366, | |
| "eval_samples_per_second": 63.846, | |
| "eval_steps_per_second": 7.991, | |
| "step": 1542 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1.675745784695201e-05, | |
| "loss": 2.4094, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 1.5946822308690015e-05, | |
| "loss": 2.3802, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 1.5136186770428017e-05, | |
| "loss": 2.3622, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.0509471893310547, | |
| "eval_runtime": 81.3625, | |
| "eval_samples_per_second": 63.825, | |
| "eval_steps_per_second": 7.989, | |
| "step": 3084 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 1.432555123216602e-05, | |
| "loss": 2.3547, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 1.3514915693904023e-05, | |
| "loss": 2.3319, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 1.2704280155642024e-05, | |
| "loss": 2.3303, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.0795907974243164, | |
| "eval_runtime": 81.2667, | |
| "eval_samples_per_second": 63.901, | |
| "eval_steps_per_second": 7.998, | |
| "step": 4626 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 1.1893644617380028e-05, | |
| "loss": 2.293, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 1.108300907911803e-05, | |
| "loss": 2.2819, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 1.027237354085603e-05, | |
| "loss": 2.2865, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.9920992851257324, | |
| "eval_runtime": 81.5086, | |
| "eval_samples_per_second": 63.711, | |
| "eval_steps_per_second": 7.975, | |
| "step": 6168 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 9.461738002594033e-06, | |
| "loss": 2.2856, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "learning_rate": 8.651102464332037e-06, | |
| "loss": 2.2669, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "learning_rate": 7.84046692607004e-06, | |
| "loss": 2.2598, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.9843554496765137, | |
| "eval_runtime": 81.4242, | |
| "eval_samples_per_second": 63.777, | |
| "eval_steps_per_second": 7.983, | |
| "step": 7710 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "learning_rate": 7.029831387808041e-06, | |
| "loss": 2.2482, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "learning_rate": 6.219195849546045e-06, | |
| "loss": 2.2285, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "learning_rate": 5.4085603112840465e-06, | |
| "loss": 2.2544, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 2.0076751708984375, | |
| "eval_runtime": 81.3076, | |
| "eval_samples_per_second": 63.869, | |
| "eval_steps_per_second": 7.994, | |
| "step": 9252 | |
| }, | |
| { | |
| "epoch": 6.16, | |
| "learning_rate": 4.5979247730220496e-06, | |
| "loss": 2.2357, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "learning_rate": 3.7872892347600522e-06, | |
| "loss": 2.218, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "learning_rate": 2.9766536964980545e-06, | |
| "loss": 2.2128, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.9913570880889893, | |
| "eval_runtime": 81.2325, | |
| "eval_samples_per_second": 63.928, | |
| "eval_steps_per_second": 8.002, | |
| "step": 10794 | |
| }, | |
| { | |
| "epoch": 7.13, | |
| "learning_rate": 2.166018158236057e-06, | |
| "loss": 2.2075, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 7.46, | |
| "learning_rate": 1.3553826199740596e-06, | |
| "loss": 2.1936, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 5.447470817120623e-07, | |
| "loss": 2.1963, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.9882222414016724, | |
| "eval_runtime": 81.2899, | |
| "eval_samples_per_second": 63.883, | |
| "eval_steps_per_second": 7.996, | |
| "step": 12336 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 12336, | |
| "total_flos": 2.0776857912786125e+17, | |
| "train_loss": 2.304505779894719, | |
| "train_runtime": 46564.1697, | |
| "train_samples_per_second": 16.953, | |
| "train_steps_per_second": 0.265 | |
| } | |
| ], | |
| "max_steps": 12336, | |
| "num_train_epochs": 8, | |
| "total_flos": 2.0776857912786125e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |