| { | |
| "best_metric": 0.8037025332450867, | |
| "best_model_checkpoint": "/mnt/bn/qingyi-bn-lq/llama/saved-alpaca-30b/checkpoint-1000", | |
| "epoch": 2.5624599615631007, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 5.9999999999999995e-05, | |
| "loss": 1.9598, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 1.372, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 0.9191, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 0.8468, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8366, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00029439252336448596, | |
| "loss": 0.819, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00028878504672897194, | |
| "loss": 0.8205, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0002831775700934579, | |
| "loss": 0.8168, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0002775700934579439, | |
| "loss": 0.8019, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002719626168224299, | |
| "loss": 0.8035, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 0.8181663751602173, | |
| "eval_runtime": 27.5256, | |
| "eval_samples_per_second": 72.66, | |
| "eval_steps_per_second": 1.163, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00026635514018691586, | |
| "loss": 0.8029, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00026074766355140184, | |
| "loss": 0.8026, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0002551401869158878, | |
| "loss": 0.8023, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.0002495327102803738, | |
| "loss": 0.8077, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.0002439252336448598, | |
| "loss": 0.7959, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.00023831775700934577, | |
| "loss": 0.797, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.00023271028037383175, | |
| "loss": 0.7937, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.00022710280373831773, | |
| "loss": 0.7939, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.0002214953271028037, | |
| "loss": 0.7904, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002158878504672897, | |
| "loss": 0.7992, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 0.8100255131721497, | |
| "eval_runtime": 27.5094, | |
| "eval_samples_per_second": 72.702, | |
| "eval_steps_per_second": 1.163, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.00021028037383177567, | |
| "loss": 0.7886, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 0.00020467289719626166, | |
| "loss": 0.7959, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.00019906542056074764, | |
| "loss": 0.7736, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.00019345794392523362, | |
| "loss": 0.7814, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.0001878504672897196, | |
| "loss": 0.7876, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.00018224299065420558, | |
| "loss": 0.7823, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.00017663551401869156, | |
| "loss": 0.7913, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.00017102803738317754, | |
| "loss": 0.7804, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 0.00016542056074766352, | |
| "loss": 0.7902, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.0001598130841121495, | |
| "loss": 0.7916, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "eval_loss": 0.8062512278556824, | |
| "eval_runtime": 27.5287, | |
| "eval_samples_per_second": 72.651, | |
| "eval_steps_per_second": 1.162, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 0.0001542056074766355, | |
| "loss": 0.7736, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 0.00014859813084112147, | |
| "loss": 0.7774, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00014299065420560745, | |
| "loss": 0.7783, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 0.00013738317757009343, | |
| "loss": 0.7644, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.0001317757009345794, | |
| "loss": 0.776, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 0.0001261682242990654, | |
| "loss": 0.7763, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 0.00012056074766355139, | |
| "loss": 0.7737, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 0.00011495327102803737, | |
| "loss": 0.7834, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.00010934579439252335, | |
| "loss": 0.7858, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 0.00010373831775700933, | |
| "loss": 0.7706, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "eval_loss": 0.8045574426651001, | |
| "eval_runtime": 27.455, | |
| "eval_samples_per_second": 72.846, | |
| "eval_steps_per_second": 1.166, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 9.813084112149531e-05, | |
| "loss": 0.7733, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 9.25233644859813e-05, | |
| "loss": 0.7761, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 8.691588785046728e-05, | |
| "loss": 0.7811, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 8.130841121495326e-05, | |
| "loss": 0.7734, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 7.570093457943924e-05, | |
| "loss": 0.7768, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 7.009345794392522e-05, | |
| "loss": 0.7714, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 6.44859813084112e-05, | |
| "loss": 0.7679, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 5.887850467289719e-05, | |
| "loss": 0.776, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 5.327102803738317e-05, | |
| "loss": 0.7705, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 4.766355140186915e-05, | |
| "loss": 0.7798, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "eval_loss": 0.8037025332450867, | |
| "eval_runtime": 27.5024, | |
| "eval_samples_per_second": 72.721, | |
| "eval_steps_per_second": 1.164, | |
| "step": 1000 | |
| } | |
| ], | |
| "max_steps": 1170, | |
| "num_train_epochs": 3, | |
| "total_flos": 6.35529279470515e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |