| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 8.0, | |
| "global_step": 952, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 9.894957983193278e-05, | |
| "loss": 0.5818, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 9.789915966386555e-05, | |
| "loss": 0.4739, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 9.684873949579833e-05, | |
| "loss": 0.4468, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 9.579831932773111e-05, | |
| "loss": 0.4327, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 9.474789915966387e-05, | |
| "loss": 0.4303, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 9.369747899159664e-05, | |
| "loss": 0.4184, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 9.264705882352942e-05, | |
| "loss": 0.4001, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 9.159663865546218e-05, | |
| "loss": 0.3856, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 9.054621848739496e-05, | |
| "loss": 0.3843, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 8.949579831932774e-05, | |
| "loss": 0.3751, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_loss": 0.3845197769571269, | |
| "eval_runtime": 3.5352, | |
| "eval_samples_per_second": 121.067, | |
| "eval_steps_per_second": 15.275, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 8.84453781512605e-05, | |
| "loss": 0.3852, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 8.739495798319329e-05, | |
| "loss": 0.3678, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 8.634453781512606e-05, | |
| "loss": 0.3489, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 8.529411764705883e-05, | |
| "loss": 0.3346, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 8.42436974789916e-05, | |
| "loss": 0.3417, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 8.319327731092437e-05, | |
| "loss": 0.334, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 8.214285714285714e-05, | |
| "loss": 0.3318, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 8.109243697478992e-05, | |
| "loss": 0.32, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 8.00420168067227e-05, | |
| "loss": 0.3125, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 7.899159663865546e-05, | |
| "loss": 0.3237, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_loss": 0.32147175100431824, | |
| "eval_runtime": 4.3808, | |
| "eval_samples_per_second": 97.7, | |
| "eval_steps_per_second": 12.327, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 7.794117647058824e-05, | |
| "loss": 0.3103, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 7.689075630252102e-05, | |
| "loss": 0.3079, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 7.584033613445379e-05, | |
| "loss": 0.3046, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 7.478991596638657e-05, | |
| "loss": 0.3002, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 7.373949579831933e-05, | |
| "loss": 0.2845, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 7.26890756302521e-05, | |
| "loss": 0.2738, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 7.163865546218488e-05, | |
| "loss": 0.2752, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 7.058823529411765e-05, | |
| "loss": 0.2737, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 6.953781512605042e-05, | |
| "loss": 0.2757, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 6.84873949579832e-05, | |
| "loss": 0.2812, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "eval_loss": 0.30918010934225554, | |
| "eval_runtime": 4.0582, | |
| "eval_samples_per_second": 105.466, | |
| "eval_steps_per_second": 13.306, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 6.743697478991598e-05, | |
| "loss": 0.2727, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 6.638655462184874e-05, | |
| "loss": 0.27, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 6.533613445378152e-05, | |
| "loss": 0.2669, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 6.428571428571429e-05, | |
| "loss": 0.2613, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 6.323529411764705e-05, | |
| "loss": 0.2671, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 6.218487394957983e-05, | |
| "loss": 0.2649, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 6.113445378151261e-05, | |
| "loss": 0.249, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 6.008403361344538e-05, | |
| "loss": 0.2442, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 5.9033613445378156e-05, | |
| "loss": 0.2476, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 5.7983193277310935e-05, | |
| "loss": 0.2441, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "eval_loss": 0.29600601734145254, | |
| "eval_runtime": 4.0145, | |
| "eval_samples_per_second": 106.614, | |
| "eval_steps_per_second": 13.451, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 5.69327731092437e-05, | |
| "loss": 0.2413, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 5.588235294117647e-05, | |
| "loss": 0.252, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 5.483193277310925e-05, | |
| "loss": 0.2421, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 5.378151260504202e-05, | |
| "loss": 0.2556, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 5.2731092436974796e-05, | |
| "loss": 0.2351, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 5.168067226890757e-05, | |
| "loss": 0.2447, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 5.0630252100840334e-05, | |
| "loss": 0.2365, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 4.957983193277311e-05, | |
| "loss": 0.2312, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 4.8529411764705885e-05, | |
| "loss": 0.2276, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "learning_rate": 4.747899159663866e-05, | |
| "loss": 0.2268, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "eval_loss": 0.28844825865122026, | |
| "eval_runtime": 4.1301, | |
| "eval_samples_per_second": 103.629, | |
| "eval_steps_per_second": 13.075, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "learning_rate": 4.642857142857143e-05, | |
| "loss": 0.2269, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 4.53781512605042e-05, | |
| "loss": 0.2307, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "learning_rate": 4.4327731092436974e-05, | |
| "loss": 0.2209, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "learning_rate": 4.327731092436975e-05, | |
| "loss": 0.2231, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 4.2226890756302525e-05, | |
| "loss": 0.2276, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "learning_rate": 4.11764705882353e-05, | |
| "loss": 0.2215, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 4.012605042016807e-05, | |
| "loss": 0.2159, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 3.907563025210084e-05, | |
| "loss": 0.2256, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "learning_rate": 3.8025210084033614e-05, | |
| "loss": 0.2209, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 3.697478991596639e-05, | |
| "loss": 0.2236, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "eval_loss": 0.28733093379126323, | |
| "eval_runtime": 4.0382, | |
| "eval_samples_per_second": 105.987, | |
| "eval_steps_per_second": 13.372, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.13, | |
| "learning_rate": 3.592436974789916e-05, | |
| "loss": 0.2134, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "learning_rate": 3.487394957983193e-05, | |
| "loss": 0.2092, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 5.29, | |
| "learning_rate": 3.382352941176471e-05, | |
| "loss": 0.2103, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 3.277310924369748e-05, | |
| "loss": 0.2131, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "learning_rate": 3.1722689075630254e-05, | |
| "loss": 0.2091, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.55, | |
| "learning_rate": 3.0672268907563026e-05, | |
| "loss": 0.2128, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 5.63, | |
| "learning_rate": 2.96218487394958e-05, | |
| "loss": 0.2085, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 0.207, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "learning_rate": 2.7521008403361346e-05, | |
| "loss": 0.2003, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "learning_rate": 2.647058823529412e-05, | |
| "loss": 0.2027, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "eval_loss": 0.2859314660902927, | |
| "eval_runtime": 4.3959, | |
| "eval_samples_per_second": 97.363, | |
| "eval_steps_per_second": 12.284, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "learning_rate": 2.542016806722689e-05, | |
| "loss": 0.2073, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 2.4369747899159663e-05, | |
| "loss": 0.2119, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "learning_rate": 2.331932773109244e-05, | |
| "loss": 0.1967, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "learning_rate": 2.226890756302521e-05, | |
| "loss": 0.199, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 6.3, | |
| "learning_rate": 2.1218487394957983e-05, | |
| "loss": 0.2012, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "learning_rate": 2.016806722689076e-05, | |
| "loss": 0.195, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 6.47, | |
| "learning_rate": 1.9117647058823528e-05, | |
| "loss": 0.1952, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "learning_rate": 1.8067226890756303e-05, | |
| "loss": 0.1985, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.64, | |
| "learning_rate": 1.7016806722689076e-05, | |
| "loss": 0.1947, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "learning_rate": 1.5966386554621848e-05, | |
| "loss": 0.2024, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "eval_loss": 0.2877343659552041, | |
| "eval_runtime": 3.8576, | |
| "eval_samples_per_second": 110.95, | |
| "eval_steps_per_second": 13.998, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "learning_rate": 1.4915966386554622e-05, | |
| "loss": 0.2005, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 6.89, | |
| "learning_rate": 1.3865546218487396e-05, | |
| "loss": 0.1978, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "learning_rate": 1.2815126050420168e-05, | |
| "loss": 0.1956, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 1.1764705882352942e-05, | |
| "loss": 0.2002, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 1.0714285714285714e-05, | |
| "loss": 0.1944, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 7.23, | |
| "learning_rate": 9.663865546218488e-06, | |
| "loss": 0.1948, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "learning_rate": 8.61344537815126e-06, | |
| "loss": 0.1934, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 7.39, | |
| "learning_rate": 7.563025210084033e-06, | |
| "loss": 0.1954, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 6.512605042016807e-06, | |
| "loss": 0.1972, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "learning_rate": 5.46218487394958e-06, | |
| "loss": 0.1952, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "eval_loss": 0.28674467348030147, | |
| "eval_runtime": 3.9339, | |
| "eval_samples_per_second": 108.799, | |
| "eval_steps_per_second": 13.727, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "learning_rate": 4.411764705882353e-06, | |
| "loss": 0.197, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 7.73, | |
| "learning_rate": 3.361344537815126e-06, | |
| "loss": 0.1899, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 7.82, | |
| "learning_rate": 2.3109243697478996e-06, | |
| "loss": 0.192, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 7.9, | |
| "learning_rate": 1.2605042016806724e-06, | |
| "loss": 0.1943, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 7.98, | |
| "learning_rate": 2.1008403361344538e-07, | |
| "loss": 0.1908, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 952, | |
| "total_flos": 8.103143399735624e+17, | |
| "train_loss": 0.2614198288559413, | |
| "train_runtime": 441.6283, | |
| "train_samples_per_second": 23.676, | |
| "train_steps_per_second": 2.156 | |
| } | |
| ], | |
| "max_steps": 952, | |
| "num_train_epochs": 8, | |
| "total_flos": 8.103143399735624e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |