| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.0, |
| "eval_steps": 500, |
| "global_step": 750, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.08, |
| "learning_rate": 0.0004991232148123761, |
| "loss": 4.3712, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.16, |
| "learning_rate": 0.0004964990092676262, |
| "loss": 3.7266, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.24, |
| "learning_rate": 0.0004921457902821578, |
| "loss": 3.5415, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.32, |
| "learning_rate": 0.0004860940925593703, |
| "loss": 3.362, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.4, |
| "learning_rate": 0.0004783863644106502, |
| "loss": 2.9872, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.48, |
| "learning_rate": 0.0004690766700109659, |
| "loss": 2.848, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.56, |
| "learning_rate": 0.00045823031017752484, |
| "loss": 2.4691, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.64, |
| "learning_rate": 0.00044592336433145995, |
| "loss": 2.4321, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.72, |
| "learning_rate": 0.00043224215685535287, |
| "loss": 2.1654, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.8, |
| "learning_rate": 0.0004172826515897146, |
| "loss": 1.8459, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.88, |
| "learning_rate": 0.00040114977871559375, |
| "loss": 1.7236, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.96, |
| "learning_rate": 0.00038395669874474915, |
| "loss": 1.42, |
| "step": 240 |
| }, |
| { |
| "epoch": 1.04, |
| "learning_rate": 0.00036582400877996547, |
| "loss": 1.2365, |
| "step": 260 |
| }, |
| { |
| "epoch": 1.12, |
| "learning_rate": 0.00034687889661302575, |
| "loss": 0.8967, |
| "step": 280 |
| }, |
| { |
| "epoch": 1.2, |
| "learning_rate": 0.00032725424859373687, |
| "loss": 0.8539, |
| "step": 300 |
| }, |
| { |
| "epoch": 1.28, |
| "learning_rate": 0.00030708771752766396, |
| "loss": 0.8967, |
| "step": 320 |
| }, |
| { |
| "epoch": 1.36, |
| "learning_rate": 0.00028652075714060294, |
| "loss": 0.6522, |
| "step": 340 |
| }, |
| { |
| "epoch": 1.44, |
| "learning_rate": 0.0002656976298823284, |
| "loss": 0.6262, |
| "step": 360 |
| }, |
| { |
| "epoch": 1.52, |
| "learning_rate": 0.00024476439502916077, |
| "loss": 0.6353, |
| "step": 380 |
| }, |
| { |
| "epoch": 1.6, |
| "learning_rate": 0.00022386788418308668, |
| "loss": 0.4705, |
| "step": 400 |
| }, |
| { |
| "epoch": 1.68, |
| "learning_rate": 0.0002031546713535688, |
| "loss": 0.4861, |
| "step": 420 |
| }, |
| { |
| "epoch": 1.76, |
| "learning_rate": 0.00018277004484618358, |
| "loss": 0.4526, |
| "step": 440 |
| }, |
| { |
| "epoch": 1.84, |
| "learning_rate": 0.00016285698816954626, |
| "loss": 0.3909, |
| "step": 460 |
| }, |
| { |
| "epoch": 1.92, |
| "learning_rate": 0.00014355517710873183, |
| "loss": 0.3611, |
| "step": 480 |
| }, |
| { |
| "epoch": 2.0, |
| "learning_rate": 0.00012500000000000006, |
| "loss": 0.423, |
| "step": 500 |
| }, |
| { |
| "epoch": 2.08, |
| "learning_rate": 0.0001073216080788921, |
| "loss": 0.2621, |
| "step": 520 |
| }, |
| { |
| "epoch": 2.16, |
| "learning_rate": 9.064400256282756e-05, |
| "loss": 0.2386, |
| "step": 540 |
| }, |
| { |
| "epoch": 2.24, |
| "learning_rate": 7.508416487165862e-05, |
| "loss": 0.2187, |
| "step": 560 |
| }, |
| { |
| "epoch": 2.32, |
| "learning_rate": 6.075123608706093e-05, |
| "loss": 0.2298, |
| "step": 580 |
| }, |
| { |
| "epoch": 2.4, |
| "learning_rate": 4.7745751406263163e-05, |
| "loss": 0.21, |
| "step": 600 |
| }, |
| { |
| "epoch": 2.48, |
| "learning_rate": 3.615893495987335e-05, |
| "loss": 0.1537, |
| "step": 620 |
| }, |
| { |
| "epoch": 2.56, |
| "learning_rate": 2.6072059940146776e-05, |
| "loss": 0.2061, |
| "step": 640 |
| }, |
| { |
| "epoch": 2.64, |
| "learning_rate": 1.7555878527937163e-05, |
| "loss": 0.1574, |
| "step": 660 |
| }, |
| { |
| "epoch": 2.72, |
| "learning_rate": 1.067012561698319e-05, |
| "loss": 0.2123, |
| "step": 680 |
| }, |
| { |
| "epoch": 2.8, |
| "learning_rate": 5.463099816548578e-06, |
| "loss": 0.2079, |
| "step": 700 |
| }, |
| { |
| "epoch": 2.88, |
| "learning_rate": 1.9713246713805587e-06, |
| "loss": 0.1966, |
| "step": 720 |
| }, |
| { |
| "epoch": 2.96, |
| "learning_rate": 2.1929247528540418e-07, |
| "loss": 0.1839, |
| "step": 740 |
| }, |
| { |
| "epoch": 3.0, |
| "step": 750, |
| "total_flos": 4.158994380698419e+16, |
| "train_loss": 1.1695097732543944, |
| "train_runtime": 966.7484, |
| "train_samples_per_second": 3.103, |
| "train_steps_per_second": 0.776 |
| } |
| ], |
| "logging_steps": 20, |
| "max_steps": 750, |
| "num_train_epochs": 3, |
| "save_steps": 1000, |
| "total_flos": 4.158994380698419e+16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
