| { | |
| "best_metric": 0.8062512278556824, | |
| "best_model_checkpoint": "/mnt/bn/qingyi-bn-lq/llama/saved-alpaca-30b/checkpoint-600", | |
| "epoch": 1.5374759769378603, | |
| "global_step": 600, | |
| "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 | |
| } | |
| ], | |
| "max_steps": 1170, | |
| "num_train_epochs": 3, | |
| "total_flos": 3.813255128564105e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |