| { | |
| "best_metric": 0.8493452668190002, | |
| "best_model_checkpoint": "/mnt/bn/qingyi-bn-lq/llama/lora-alpaca/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": 1.9913, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 1.5044, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 0.98, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 0.9036, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8966, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00029439252336448596, | |
| "loss": 0.8743, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00028878504672897194, | |
| "loss": 0.8711, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0002831775700934579, | |
| "loss": 0.8505, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0002775700934579439, | |
| "loss": 0.8546, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002719626168224299, | |
| "loss": 0.8521, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 0.8707928657531738, | |
| "eval_runtime": 19.2615, | |
| "eval_samples_per_second": 103.834, | |
| "eval_steps_per_second": 3.271, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00026635514018691586, | |
| "loss": 0.8523, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00026074766355140184, | |
| "loss": 0.8463, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0002551401869158878, | |
| "loss": 0.8443, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.0002495327102803738, | |
| "loss": 0.8557, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.0002439252336448598, | |
| "loss": 0.8607, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.00023831775700934577, | |
| "loss": 0.8479, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.00023271028037383175, | |
| "loss": 0.8416, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.00022710280373831773, | |
| "loss": 0.8494, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.0002214953271028037, | |
| "loss": 0.8414, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002158878504672897, | |
| "loss": 0.8476, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 0.8598244190216064, | |
| "eval_runtime": 74.9831, | |
| "eval_samples_per_second": 26.673, | |
| "eval_steps_per_second": 0.84, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.00021028037383177567, | |
| "loss": 0.8443, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 0.00020467289719626166, | |
| "loss": 0.8382, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.00019906542056074764, | |
| "loss": 0.8467, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.00019345794392523362, | |
| "loss": 0.8384, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.0001878504672897196, | |
| "loss": 0.8346, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.00018224299065420558, | |
| "loss": 0.8347, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.00017663551401869156, | |
| "loss": 0.8363, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.00017102803738317754, | |
| "loss": 0.8284, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 0.00016542056074766352, | |
| "loss": 0.837, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.0001598130841121495, | |
| "loss": 0.8385, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "eval_loss": 0.8547194600105286, | |
| "eval_runtime": 19.3656, | |
| "eval_samples_per_second": 103.276, | |
| "eval_steps_per_second": 3.253, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 0.0001542056074766355, | |
| "loss": 0.8359, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 0.00014859813084112147, | |
| "loss": 0.8315, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00014299065420560745, | |
| "loss": 0.8379, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 0.00013738317757009343, | |
| "loss": 0.8372, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.0001317757009345794, | |
| "loss": 0.8277, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 0.0001261682242990654, | |
| "loss": 0.8321, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 0.00012056074766355139, | |
| "loss": 0.8299, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 0.00011495327102803737, | |
| "loss": 0.8289, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.00010934579439252335, | |
| "loss": 0.8255, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 0.00010373831775700933, | |
| "loss": 0.8319, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "eval_loss": 0.8514933586120605, | |
| "eval_runtime": 19.1889, | |
| "eval_samples_per_second": 104.227, | |
| "eval_steps_per_second": 3.283, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 9.813084112149531e-05, | |
| "loss": 0.8241, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 9.25233644859813e-05, | |
| "loss": 0.8214, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 8.691588785046728e-05, | |
| "loss": 0.828, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 8.130841121495326e-05, | |
| "loss": 0.8263, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 7.570093457943924e-05, | |
| "loss": 0.8275, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 7.009345794392522e-05, | |
| "loss": 0.8277, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 6.44859813084112e-05, | |
| "loss": 0.8315, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 5.887850467289719e-05, | |
| "loss": 0.8239, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 5.327102803738317e-05, | |
| "loss": 0.8243, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 4.766355140186915e-05, | |
| "loss": 0.8189, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "eval_loss": 0.8493452668190002, | |
| "eval_runtime": 19.4174, | |
| "eval_samples_per_second": 103.0, | |
| "eval_steps_per_second": 3.245, | |
| "step": 1000 | |
| } | |
| ], | |
| "max_steps": 1170, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.299718813859709e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |