| { |
| "best_metric": 1.388809323310852, |
| "best_model_checkpoint": "./modelversion01/checkpoint-600", |
| "epoch": 1.0, |
| "global_step": 639, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.02, |
| "learning_rate": 0.0001968701095461659, |
| "loss": 3.6977, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.03, |
| "learning_rate": 0.00019374021909233179, |
| "loss": 3.7426, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.05, |
| "learning_rate": 0.00019061032863849765, |
| "loss": 3.5577, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.06, |
| "learning_rate": 0.00018748043818466354, |
| "loss": 3.3304, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.08, |
| "learning_rate": 0.00018435054773082942, |
| "loss": 3.3761, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.09, |
| "learning_rate": 0.0001812206572769953, |
| "loss": 3.2324, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.11, |
| "learning_rate": 0.0001780907668231612, |
| "loss": 3.1829, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.13, |
| "learning_rate": 0.0001749608763693271, |
| "loss": 3.0992, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.14, |
| "learning_rate": 0.00017183098591549295, |
| "loss": 3.2535, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.16, |
| "learning_rate": 0.00016870109546165884, |
| "loss": 3.1304, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_accuracy": 0.3253662297609869, |
| "eval_loss": 3.154646873474121, |
| "eval_runtime": 30.5027, |
| "eval_samples_per_second": 85.042, |
| "eval_steps_per_second": 10.655, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.17, |
| "learning_rate": 0.00016557120500782473, |
| "loss": 3.1477, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.19, |
| "learning_rate": 0.00016244131455399061, |
| "loss": 2.9492, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.2, |
| "learning_rate": 0.0001593114241001565, |
| "loss": 2.7843, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.22, |
| "learning_rate": 0.00015618153364632236, |
| "loss": 2.919, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.23, |
| "learning_rate": 0.00015305164319248828, |
| "loss": 2.703, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.25, |
| "learning_rate": 0.00014992175273865417, |
| "loss": 2.7079, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.27, |
| "learning_rate": 0.00014679186228482006, |
| "loss": 2.5964, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.28, |
| "learning_rate": 0.00014366197183098594, |
| "loss": 2.7854, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.3, |
| "learning_rate": 0.0001405320813771518, |
| "loss": 2.4903, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.31, |
| "learning_rate": 0.0001374021909233177, |
| "loss": 2.6514, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.31, |
| "eval_accuracy": 0.4853508095605243, |
| "eval_loss": 2.5057504177093506, |
| "eval_runtime": 34.1479, |
| "eval_samples_per_second": 75.964, |
| "eval_steps_per_second": 9.517, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.33, |
| "learning_rate": 0.00013427230046948358, |
| "loss": 2.4657, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.34, |
| "learning_rate": 0.00013114241001564947, |
| "loss": 2.4503, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.36, |
| "learning_rate": 0.00012801251956181536, |
| "loss": 2.2323, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.38, |
| "learning_rate": 0.00012488262910798122, |
| "loss": 2.1866, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.39, |
| "learning_rate": 0.00012175273865414711, |
| "loss": 2.3176, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.41, |
| "learning_rate": 0.000118622848200313, |
| "loss": 2.2193, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.42, |
| "learning_rate": 0.00011549295774647888, |
| "loss": 2.1877, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.44, |
| "learning_rate": 0.00011236306729264476, |
| "loss": 2.2086, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.45, |
| "learning_rate": 0.00010923317683881065, |
| "loss": 2.2096, |
| "step": 290 |
| }, |
| { |
| "epoch": 0.47, |
| "learning_rate": 0.00010610328638497653, |
| "loss": 2.0636, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.47, |
| "eval_accuracy": 0.5771010023130301, |
| "eval_loss": 2.0646958351135254, |
| "eval_runtime": 34.5408, |
| "eval_samples_per_second": 75.1, |
| "eval_steps_per_second": 9.409, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.49, |
| "learning_rate": 0.00010297339593114241, |
| "loss": 2.1629, |
| "step": 310 |
| }, |
| { |
| "epoch": 0.5, |
| "learning_rate": 9.98435054773083e-05, |
| "loss": 1.9858, |
| "step": 320 |
| }, |
| { |
| "epoch": 0.52, |
| "learning_rate": 9.671361502347419e-05, |
| "loss": 2.1714, |
| "step": 330 |
| }, |
| { |
| "epoch": 0.53, |
| "learning_rate": 9.358372456964006e-05, |
| "loss": 1.9333, |
| "step": 340 |
| }, |
| { |
| "epoch": 0.55, |
| "learning_rate": 9.045383411580595e-05, |
| "loss": 1.9808, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.56, |
| "learning_rate": 8.732394366197182e-05, |
| "loss": 1.8507, |
| "step": 360 |
| }, |
| { |
| "epoch": 0.58, |
| "learning_rate": 8.419405320813773e-05, |
| "loss": 1.9109, |
| "step": 370 |
| }, |
| { |
| "epoch": 0.59, |
| "learning_rate": 8.106416275430361e-05, |
| "loss": 1.7645, |
| "step": 380 |
| }, |
| { |
| "epoch": 0.61, |
| "learning_rate": 7.793427230046949e-05, |
| "loss": 1.7828, |
| "step": 390 |
| }, |
| { |
| "epoch": 0.63, |
| "learning_rate": 7.480438184663538e-05, |
| "loss": 1.7812, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.63, |
| "eval_accuracy": 0.6422513492675405, |
| "eval_loss": 1.7535737752914429, |
| "eval_runtime": 34.377, |
| "eval_samples_per_second": 75.457, |
| "eval_steps_per_second": 9.454, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "learning_rate": 7.167449139280125e-05, |
| "loss": 1.7357, |
| "step": 410 |
| }, |
| { |
| "epoch": 0.66, |
| "learning_rate": 6.854460093896714e-05, |
| "loss": 1.8218, |
| "step": 420 |
| }, |
| { |
| "epoch": 0.67, |
| "learning_rate": 6.541471048513303e-05, |
| "loss": 1.504, |
| "step": 430 |
| }, |
| { |
| "epoch": 0.69, |
| "learning_rate": 6.22848200312989e-05, |
| "loss": 1.665, |
| "step": 440 |
| }, |
| { |
| "epoch": 0.7, |
| "learning_rate": 5.915492957746479e-05, |
| "loss": 1.6636, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.72, |
| "learning_rate": 5.602503912363067e-05, |
| "loss": 1.7505, |
| "step": 460 |
| }, |
| { |
| "epoch": 0.74, |
| "learning_rate": 5.2895148669796554e-05, |
| "loss": 1.528, |
| "step": 470 |
| }, |
| { |
| "epoch": 0.75, |
| "learning_rate": 4.976525821596245e-05, |
| "loss": 1.4391, |
| "step": 480 |
| }, |
| { |
| "epoch": 0.77, |
| "learning_rate": 4.663536776212833e-05, |
| "loss": 1.6142, |
| "step": 490 |
| }, |
| { |
| "epoch": 0.78, |
| "learning_rate": 4.350547730829421e-05, |
| "loss": 1.5857, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.78, |
| "eval_accuracy": 0.6973785659213569, |
| "eval_loss": 1.5272208452224731, |
| "eval_runtime": 33.0601, |
| "eval_samples_per_second": 78.463, |
| "eval_steps_per_second": 9.831, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "learning_rate": 4.037558685446009e-05, |
| "loss": 1.5904, |
| "step": 510 |
| }, |
| { |
| "epoch": 0.81, |
| "learning_rate": 3.724569640062598e-05, |
| "loss": 1.4461, |
| "step": 520 |
| }, |
| { |
| "epoch": 0.83, |
| "learning_rate": 3.411580594679186e-05, |
| "loss": 1.5556, |
| "step": 530 |
| }, |
| { |
| "epoch": 0.85, |
| "learning_rate": 3.0985915492957744e-05, |
| "loss": 1.6289, |
| "step": 540 |
| }, |
| { |
| "epoch": 0.86, |
| "learning_rate": 2.7856025039123636e-05, |
| "loss": 1.4632, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.88, |
| "learning_rate": 2.4726134585289514e-05, |
| "loss": 1.3283, |
| "step": 560 |
| }, |
| { |
| "epoch": 0.89, |
| "learning_rate": 2.1596244131455402e-05, |
| "loss": 1.3679, |
| "step": 570 |
| }, |
| { |
| "epoch": 0.91, |
| "learning_rate": 1.8466353677621284e-05, |
| "loss": 1.4145, |
| "step": 580 |
| }, |
| { |
| "epoch": 0.92, |
| "learning_rate": 1.533646322378717e-05, |
| "loss": 1.3822, |
| "step": 590 |
| }, |
| { |
| "epoch": 0.94, |
| "learning_rate": 1.2206572769953052e-05, |
| "loss": 1.3055, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.94, |
| "eval_accuracy": 0.7224363916730917, |
| "eval_loss": 1.388809323310852, |
| "eval_runtime": 29.6379, |
| "eval_samples_per_second": 87.523, |
| "eval_steps_per_second": 10.966, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.95, |
| "learning_rate": 9.076682316118937e-06, |
| "loss": 1.4099, |
| "step": 610 |
| }, |
| { |
| "epoch": 0.97, |
| "learning_rate": 5.94679186228482e-06, |
| "loss": 1.3346, |
| "step": 620 |
| }, |
| { |
| "epoch": 0.99, |
| "learning_rate": 2.8169014084507042e-06, |
| "loss": 1.4532, |
| "step": 630 |
| }, |
| { |
| "epoch": 1.0, |
| "step": 639, |
| "total_flos": 7.922356224994714e+17, |
| "train_loss": 2.18192570049065, |
| "train_runtime": 471.371, |
| "train_samples_per_second": 21.66, |
| "train_steps_per_second": 1.356 |
| } |
| ], |
| "max_steps": 639, |
| "num_train_epochs": 1, |
| "total_flos": 7.922356224994714e+17, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
