| { | |
| "best_metric": 0.8164276480674744, | |
| "best_model_checkpoint": "/mnt/bn/qingyi-bn-lq/llama/saved-alpaca-13b/checkpoint-800", | |
| "epoch": 2.0499679692504804, | |
| "global_step": 800, | |
| "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": 1.9024, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 1.4401, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 0.9439, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 0.8693, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8598, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00029439252336448596, | |
| "loss": 0.8485, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00028878504672897194, | |
| "loss": 0.8323, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0002831775700934579, | |
| "loss": 0.8364, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0002775700934579439, | |
| "loss": 0.8364, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002719626168224299, | |
| "loss": 0.842, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 0.8336600661277771, | |
| "eval_runtime": 14.4551, | |
| "eval_samples_per_second": 138.359, | |
| "eval_steps_per_second": 2.214, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00026635514018691586, | |
| "loss": 0.8289, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00026074766355140184, | |
| "loss": 0.8383, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0002551401869158878, | |
| "loss": 0.822, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.0002495327102803738, | |
| "loss": 0.8378, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.0002439252336448598, | |
| "loss": 0.8275, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.00023831775700934577, | |
| "loss": 0.8225, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.00023271028037383175, | |
| "loss": 0.8188, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.00022710280373831773, | |
| "loss": 0.8251, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.0002214953271028037, | |
| "loss": 0.8107, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002158878504672897, | |
| "loss": 0.806, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 0.8227179050445557, | |
| "eval_runtime": 14.4306, | |
| "eval_samples_per_second": 138.594, | |
| "eval_steps_per_second": 2.218, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.00021028037383177567, | |
| "loss": 0.8157, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 0.00020467289719626166, | |
| "loss": 0.8139, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.00019906542056074764, | |
| "loss": 0.8203, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.00019345794392523362, | |
| "loss": 0.8183, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.0001878504672897196, | |
| "loss": 0.8046, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.00018224299065420558, | |
| "loss": 0.8053, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.00017663551401869156, | |
| "loss": 0.8037, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.00017102803738317754, | |
| "loss": 0.8036, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 0.00016542056074766352, | |
| "loss": 0.7971, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.0001598130841121495, | |
| "loss": 0.8024, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "eval_loss": 0.8188450932502747, | |
| "eval_runtime": 14.5635, | |
| "eval_samples_per_second": 137.329, | |
| "eval_steps_per_second": 2.197, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 0.0001542056074766355, | |
| "loss": 0.8042, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 0.00014859813084112147, | |
| "loss": 0.8148, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00014299065420560745, | |
| "loss": 0.7976, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 0.00013738317757009343, | |
| "loss": 0.8112, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.0001317757009345794, | |
| "loss": 0.805, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 0.0001261682242990654, | |
| "loss": 0.797, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 0.00012056074766355139, | |
| "loss": 0.7883, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 0.00011495327102803737, | |
| "loss": 0.8026, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.00010934579439252335, | |
| "loss": 0.8098, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 0.00010373831775700933, | |
| "loss": 0.8021, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "eval_loss": 0.8164276480674744, | |
| "eval_runtime": 14.5011, | |
| "eval_samples_per_second": 137.921, | |
| "eval_steps_per_second": 2.207, | |
| "step": 800 | |
| } | |
| ], | |
| "max_steps": 1170, | |
| "num_train_epochs": 3, | |
| "total_flos": 2.022163468739674e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |