| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.6666666666666666, | |
| "eval_steps": 106, | |
| "global_step": 702, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.000949667616334283, | |
| "eval_loss": 1.492834210395813, | |
| "eval_runtime": 241.5208, | |
| "eval_samples_per_second": 4.844, | |
| "eval_steps_per_second": 2.422, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00949667616334283, | |
| "grad_norm": 0.782672107219696, | |
| "learning_rate": 4e-08, | |
| "loss": 1.4416, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01899335232668566, | |
| "grad_norm": 0.5684090852737427, | |
| "learning_rate": 8e-08, | |
| "loss": 1.4691, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02849002849002849, | |
| "grad_norm": 0.7916907668113708, | |
| "learning_rate": 1.2e-07, | |
| "loss": 1.4888, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03798670465337132, | |
| "grad_norm": 0.7482171058654785, | |
| "learning_rate": 1.6e-07, | |
| "loss": 1.4607, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.04748338081671415, | |
| "grad_norm": 0.8853126168251038, | |
| "learning_rate": 2e-07, | |
| "loss": 1.4953, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05698005698005698, | |
| "grad_norm": 0.6968705058097839, | |
| "learning_rate": 2.4e-07, | |
| "loss": 1.4915, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.06647673314339982, | |
| "grad_norm": 0.737502932548523, | |
| "learning_rate": 2.8e-07, | |
| "loss": 1.459, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.07597340930674264, | |
| "grad_norm": 0.6606144309043884, | |
| "learning_rate": 3.2e-07, | |
| "loss": 1.4511, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.08547008547008547, | |
| "grad_norm": 0.7408589720726013, | |
| "learning_rate": 3.6e-07, | |
| "loss": 1.4256, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.0949667616334283, | |
| "grad_norm": 0.8764765858650208, | |
| "learning_rate": 4e-07, | |
| "loss": 1.476, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.100664767331434, | |
| "eval_loss": 1.491244912147522, | |
| "eval_runtime": 241.0891, | |
| "eval_samples_per_second": 4.853, | |
| "eval_steps_per_second": 2.426, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 0.10446343779677113, | |
| "grad_norm": 0.7903630137443542, | |
| "learning_rate": 4.3999999999999997e-07, | |
| "loss": 1.4327, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.11396011396011396, | |
| "grad_norm": 0.6204758286476135, | |
| "learning_rate": 4.8e-07, | |
| "loss": 1.4546, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.12345679012345678, | |
| "grad_norm": 0.6556420922279358, | |
| "learning_rate": 5.2e-07, | |
| "loss": 1.4785, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.13295346628679963, | |
| "grad_norm": 0.6578308939933777, | |
| "learning_rate": 5.6e-07, | |
| "loss": 1.4848, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.14245014245014245, | |
| "grad_norm": 0.7534553408622742, | |
| "learning_rate": 6e-07, | |
| "loss": 1.4714, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.15194681861348527, | |
| "grad_norm": 1.1248106956481934, | |
| "learning_rate": 6.4e-07, | |
| "loss": 1.4513, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.16144349477682812, | |
| "grad_norm": 0.7845984697341919, | |
| "learning_rate": 6.800000000000001e-07, | |
| "loss": 1.4494, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.17094017094017094, | |
| "grad_norm": 0.8384345769882202, | |
| "learning_rate": 7.2e-07, | |
| "loss": 1.4326, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.18043684710351376, | |
| "grad_norm": 0.737319827079773, | |
| "learning_rate": 7.599999999999999e-07, | |
| "loss": 1.4772, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.1899335232668566, | |
| "grad_norm": 0.5843039155006409, | |
| "learning_rate": 8e-07, | |
| "loss": 1.459, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.19943019943019943, | |
| "grad_norm": 0.6763444542884827, | |
| "learning_rate": 8.399999999999999e-07, | |
| "loss": 1.4464, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.201329534662868, | |
| "eval_loss": 1.4787342548370361, | |
| "eval_runtime": 241.3275, | |
| "eval_samples_per_second": 4.848, | |
| "eval_steps_per_second": 2.424, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 0.20892687559354226, | |
| "grad_norm": 0.5159311294555664, | |
| "learning_rate": 8.799999999999999e-07, | |
| "loss": 1.4781, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.2184235517568851, | |
| "grad_norm": 0.635596513748169, | |
| "learning_rate": 9.2e-07, | |
| "loss": 1.4475, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.22792022792022792, | |
| "grad_norm": 0.6518263220787048, | |
| "learning_rate": 9.6e-07, | |
| "loss": 1.4173, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.23741690408357075, | |
| "grad_norm": 0.6334572434425354, | |
| "learning_rate": 1e-06, | |
| "loss": 1.4194, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.24691358024691357, | |
| "grad_norm": 0.7460399270057678, | |
| "learning_rate": 1.04e-06, | |
| "loss": 1.4185, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.2564102564102564, | |
| "grad_norm": 0.5270726084709167, | |
| "learning_rate": 1.08e-06, | |
| "loss": 1.385, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.26590693257359926, | |
| "grad_norm": 0.5877587795257568, | |
| "learning_rate": 1.12e-06, | |
| "loss": 1.412, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.2754036087369421, | |
| "grad_norm": 0.7391476035118103, | |
| "learning_rate": 1.16e-06, | |
| "loss": 1.4478, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.2849002849002849, | |
| "grad_norm": 0.7564486265182495, | |
| "learning_rate": 1.2e-06, | |
| "loss": 1.439, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.2943969610636277, | |
| "grad_norm": 0.5443644523620605, | |
| "learning_rate": 1.24e-06, | |
| "loss": 1.4065, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.301994301994302, | |
| "eval_loss": 1.438108205795288, | |
| "eval_runtime": 241.1917, | |
| "eval_samples_per_second": 4.851, | |
| "eval_steps_per_second": 2.425, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 0.30389363722697055, | |
| "grad_norm": 0.5446246862411499, | |
| "learning_rate": 1.28e-06, | |
| "loss": 1.4128, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.31339031339031337, | |
| "grad_norm": 0.7768235802650452, | |
| "learning_rate": 1.32e-06, | |
| "loss": 1.3838, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.32288698955365625, | |
| "grad_norm": 0.8241527080535889, | |
| "learning_rate": 1.3600000000000001e-06, | |
| "loss": 1.4541, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.33238366571699907, | |
| "grad_norm": 0.7136777639389038, | |
| "learning_rate": 1.4e-06, | |
| "loss": 1.4092, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.3418803418803419, | |
| "grad_norm": 0.6339492201805115, | |
| "learning_rate": 1.44e-06, | |
| "loss": 1.4339, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.3513770180436847, | |
| "grad_norm": 0.5841691493988037, | |
| "learning_rate": 1.48e-06, | |
| "loss": 1.3396, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.36087369420702753, | |
| "grad_norm": 0.6338390707969666, | |
| "learning_rate": 1.5199999999999998e-06, | |
| "loss": 1.3738, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.37037037037037035, | |
| "grad_norm": 0.6443079710006714, | |
| "learning_rate": 1.5599999999999999e-06, | |
| "loss": 1.3632, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.3798670465337132, | |
| "grad_norm": 0.7841497659683228, | |
| "learning_rate": 1.6e-06, | |
| "loss": 1.386, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.38936372269705605, | |
| "grad_norm": 0.6738791465759277, | |
| "learning_rate": 1.6399999999999998e-06, | |
| "loss": 1.3469, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.39886039886039887, | |
| "grad_norm": 0.6674798727035522, | |
| "learning_rate": 1.6799999999999998e-06, | |
| "loss": 1.3565, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.402659069325736, | |
| "eval_loss": 1.371882438659668, | |
| "eval_runtime": 241.1901, | |
| "eval_samples_per_second": 4.851, | |
| "eval_steps_per_second": 2.425, | |
| "step": 424 | |
| }, | |
| { | |
| "epoch": 0.4083570750237417, | |
| "grad_norm": 0.7130848169326782, | |
| "learning_rate": 1.7199999999999998e-06, | |
| "loss": 1.3734, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.4178537511870845, | |
| "grad_norm": 0.60502690076828, | |
| "learning_rate": 1.7599999999999999e-06, | |
| "loss": 1.3689, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.42735042735042733, | |
| "grad_norm": 0.7991560697555542, | |
| "learning_rate": 1.8e-06, | |
| "loss": 1.335, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.4368471035137702, | |
| "grad_norm": 0.7825191020965576, | |
| "learning_rate": 1.84e-06, | |
| "loss": 1.3713, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.44634377967711303, | |
| "grad_norm": 0.8290002942085266, | |
| "learning_rate": 1.8799999999999998e-06, | |
| "loss": 1.3294, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.45584045584045585, | |
| "grad_norm": 0.7695876955986023, | |
| "learning_rate": 1.92e-06, | |
| "loss": 1.2743, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.46533713200379867, | |
| "grad_norm": 0.7531653046607971, | |
| "learning_rate": 1.96e-06, | |
| "loss": 1.3175, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.4748338081671415, | |
| "grad_norm": 0.9283966422080994, | |
| "learning_rate": 2e-06, | |
| "loss": 1.2777, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4843304843304843, | |
| "grad_norm": 0.8161536455154419, | |
| "learning_rate": 1.998386746257178e-06, | |
| "loss": 1.2737, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.49382716049382713, | |
| "grad_norm": 0.8506576418876648, | |
| "learning_rate": 1.993552190203991e-06, | |
| "loss": 1.277, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.50332383665717, | |
| "grad_norm": 0.7465077042579651, | |
| "learning_rate": 1.985511930571733e-06, | |
| "loss": 1.2351, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.50332383665717, | |
| "eval_loss": 1.285551905632019, | |
| "eval_runtime": 241.4345, | |
| "eval_samples_per_second": 4.846, | |
| "eval_steps_per_second": 2.423, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.5128205128205128, | |
| "grad_norm": 0.6605340838432312, | |
| "learning_rate": 1.9742919093182947e-06, | |
| "loss": 1.2638, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.5223171889838556, | |
| "grad_norm": 0.8043677806854248, | |
| "learning_rate": 1.959928327926239e-06, | |
| "loss": 1.2617, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.5318138651471985, | |
| "grad_norm": 0.9856230020523071, | |
| "learning_rate": 1.942467530598449e-06, | |
| "loss": 1.2799, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.5413105413105413, | |
| "grad_norm": 0.9948506355285645, | |
| "learning_rate": 1.9219658547282065e-06, | |
| "loss": 1.2449, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.5508072174738842, | |
| "grad_norm": 0.8890409469604492, | |
| "learning_rate": 1.8984894491261762e-06, | |
| "loss": 1.196, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.560303893637227, | |
| "grad_norm": 0.9632903337478638, | |
| "learning_rate": 1.872114060590769e-06, | |
| "loss": 1.2567, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.5698005698005698, | |
| "grad_norm": 1.1697239875793457, | |
| "learning_rate": 1.842924789510531e-06, | |
| "loss": 1.2497, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.5792972459639126, | |
| "grad_norm": 1.23131263256073, | |
| "learning_rate": 1.8110158152871029e-06, | |
| "loss": 1.2721, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.5887939221272555, | |
| "grad_norm": 1.192724585533142, | |
| "learning_rate": 1.776490092464676e-06, | |
| "loss": 1.1905, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.5982905982905983, | |
| "grad_norm": 0.8948692083358765, | |
| "learning_rate": 1.7394590185463837e-06, | |
| "loss": 1.1938, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.603988603988604, | |
| "eval_loss": 1.1977587938308716, | |
| "eval_runtime": 242.3209, | |
| "eval_samples_per_second": 4.828, | |
| "eval_steps_per_second": 2.414, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 0.6077872744539411, | |
| "grad_norm": 1.2288507223129272, | |
| "learning_rate": 1.7000420745694253e-06, | |
| "loss": 1.1886, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.6172839506172839, | |
| "grad_norm": 1.0895919799804688, | |
| "learning_rate": 1.6583664395986035e-06, | |
| "loss": 1.1645, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.6267806267806267, | |
| "grad_norm": 0.91950523853302, | |
| "learning_rate": 1.61456658038212e-06, | |
| "loss": 1.1593, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.6362773029439696, | |
| "grad_norm": 0.9742187857627869, | |
| "learning_rate": 1.5687838174936082e-06, | |
| "loss": 1.1584, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.6457739791073125, | |
| "grad_norm": 1.2041188478469849, | |
| "learning_rate": 1.5211658693602396e-06, | |
| "loss": 1.1502, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.6552706552706553, | |
| "grad_norm": 1.0713285207748413, | |
| "learning_rate": 1.471866375648119e-06, | |
| "loss": 1.1544, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.6647673314339981, | |
| "grad_norm": 1.3049912452697754, | |
| "learning_rate": 1.4210444015427466e-06, | |
| "loss": 1.1671, | |
| "step": 700 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1053, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 351, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.2971277740408832e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |