| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9916434540389973, | |
| "eval_steps": 200, | |
| "global_step": 807, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03714020427112349, | |
| "grad_norm": 0.9345887331215147, | |
| "learning_rate": 1.2345679012345678e-05, | |
| "loss": 0.7627, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07428040854224698, | |
| "grad_norm": 1.188973319567107, | |
| "learning_rate": 2.4691358024691357e-05, | |
| "loss": 0.7982, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.11142061281337047, | |
| "grad_norm": 0.6516884136437275, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.5529, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.14856081708449395, | |
| "grad_norm": 0.6588490543859995, | |
| "learning_rate": 4.938271604938271e-05, | |
| "loss": 0.4245, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.18570102135561745, | |
| "grad_norm": 0.4226844096352398, | |
| "learning_rate": 6.17283950617284e-05, | |
| "loss": 0.4319, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.22284122562674094, | |
| "grad_norm": 0.7485838342795699, | |
| "learning_rate": 7.407407407407407e-05, | |
| "loss": 0.3556, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.25998142989786444, | |
| "grad_norm": 0.5245389532842025, | |
| "learning_rate": 8.641975308641975e-05, | |
| "loss": 0.3381, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2971216341689879, | |
| "grad_norm": 0.5491860599786594, | |
| "learning_rate": 9.876543209876543e-05, | |
| "loss": 0.3287, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.3342618384401114, | |
| "grad_norm": 0.46666477130003114, | |
| "learning_rate": 9.996208625907141e-05, | |
| "loss": 0.2942, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3714020427112349, | |
| "grad_norm": 0.5298545126148352, | |
| "learning_rate": 9.983110023102147e-05, | |
| "loss": 0.3021, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4085422469823584, | |
| "grad_norm": 0.5132219959623447, | |
| "learning_rate": 9.960681902224691e-05, | |
| "loss": 0.3133, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.4456824512534819, | |
| "grad_norm": 0.7322228366411259, | |
| "learning_rate": 9.92896625383049e-05, | |
| "loss": 0.3027, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.4828226555246054, | |
| "grad_norm": 2.89785379394605, | |
| "learning_rate": 9.888022456844251e-05, | |
| "loss": 0.2983, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.5199628597957289, | |
| "grad_norm": 0.9602775250127539, | |
| "learning_rate": 9.837927167388792e-05, | |
| "loss": 0.2937, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.5571030640668524, | |
| "grad_norm": 0.587882249681093, | |
| "learning_rate": 9.778774175267295e-05, | |
| "loss": 0.2733, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5942432683379758, | |
| "grad_norm": 0.5952699909669362, | |
| "learning_rate": 9.710674228367423e-05, | |
| "loss": 0.3127, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.6313834726090993, | |
| "grad_norm": 0.8601590917310498, | |
| "learning_rate": 9.633754825316015e-05, | |
| "loss": 0.2846, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.6685236768802229, | |
| "grad_norm": 0.5276550802440668, | |
| "learning_rate": 9.548159976772592e-05, | |
| "loss": 0.289, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.7056638811513464, | |
| "grad_norm": 0.6256007560816419, | |
| "learning_rate": 9.454049935808568e-05, | |
| "loss": 0.2722, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.7428040854224698, | |
| "grad_norm": 0.7083045077849746, | |
| "learning_rate": 9.351600897876964e-05, | |
| "loss": 0.2724, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7428040854224698, | |
| "eval_loss": 0.32194846868515015, | |
| "eval_runtime": 69.2853, | |
| "eval_samples_per_second": 5.08, | |
| "eval_steps_per_second": 0.635, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7799442896935933, | |
| "grad_norm": 0.5489982408476731, | |
| "learning_rate": 9.241004670934348e-05, | |
| "loss": 0.2706, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.8170844939647168, | |
| "grad_norm": 0.4976435465000861, | |
| "learning_rate": 9.122468316332611e-05, | |
| "loss": 0.2878, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.8542246982358404, | |
| "grad_norm": 0.527946791379517, | |
| "learning_rate": 8.99621376115291e-05, | |
| "loss": 0.2653, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.8913649025069638, | |
| "grad_norm": 0.4489299048179587, | |
| "learning_rate": 8.862477382707568e-05, | |
| "loss": 0.2653, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.9285051067780873, | |
| "grad_norm": 0.35733720135057156, | |
| "learning_rate": 8.721509565987859e-05, | |
| "loss": 0.2613, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.9656453110492108, | |
| "grad_norm": 0.940140704929043, | |
| "learning_rate": 8.573574234886217e-05, | |
| "loss": 0.2719, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.41579714400957546, | |
| "learning_rate": 8.418948358070535e-05, | |
| "loss": 0.268, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.0371402042711235, | |
| "grad_norm": 0.44513719030149457, | |
| "learning_rate": 8.257921430435678e-05, | |
| "loss": 0.2292, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.074280408542247, | |
| "grad_norm": 0.36739314586440486, | |
| "learning_rate": 8.090794931103026e-05, | |
| "loss": 0.2161, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.1114206128133706, | |
| "grad_norm": 0.49411650365866194, | |
| "learning_rate": 7.917881758982837e-05, | |
| "loss": 0.2242, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.1485608170844939, | |
| "grad_norm": 0.4197677070916283, | |
| "learning_rate": 7.739505646956135e-05, | |
| "loss": 0.2497, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.1857010213556174, | |
| "grad_norm": 0.6822804225967691, | |
| "learning_rate": 7.556000555772967e-05, | |
| "loss": 0.2278, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.222841225626741, | |
| "grad_norm": 0.49319142186511206, | |
| "learning_rate": 7.367710048801715e-05, | |
| "loss": 0.2403, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.2599814298978644, | |
| "grad_norm": 0.5235786948264898, | |
| "learning_rate": 7.174986648800161e-05, | |
| "loss": 0.231, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.297121634168988, | |
| "grad_norm": 0.6055462763909709, | |
| "learning_rate": 6.978191177912498e-05, | |
| "loss": 0.2403, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.3342618384401115, | |
| "grad_norm": 0.5300067921269699, | |
| "learning_rate": 6.777692082128024e-05, | |
| "loss": 0.218, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.3714020427112348, | |
| "grad_norm": 0.42875365947814814, | |
| "learning_rate": 6.573864741466235e-05, | |
| "loss": 0.2291, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.4085422469823583, | |
| "grad_norm": 0.43773461983892203, | |
| "learning_rate": 6.367090767179855e-05, | |
| "loss": 0.2318, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.4456824512534818, | |
| "grad_norm": 0.44564163439616256, | |
| "learning_rate": 6.157757287291557e-05, | |
| "loss": 0.2265, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.4828226555246053, | |
| "grad_norm": 0.5057499808132443, | |
| "learning_rate": 5.946256221802051e-05, | |
| "loss": 0.2418, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.4828226555246053, | |
| "eval_loss": 0.3125605285167694, | |
| "eval_runtime": 69.1118, | |
| "eval_samples_per_second": 5.093, | |
| "eval_steps_per_second": 0.637, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.5199628597957289, | |
| "grad_norm": 1.7611100076709196, | |
| "learning_rate": 5.732983548926485e-05, | |
| "loss": 0.2205, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.5571030640668524, | |
| "grad_norm": 0.43196833026875653, | |
| "learning_rate": 5.5183385637329446e-05, | |
| "loss": 0.2239, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.594243268337976, | |
| "grad_norm": 1.0867050785435954, | |
| "learning_rate": 5.30272313057105e-05, | |
| "loss": 0.2167, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.6313834726090994, | |
| "grad_norm": 0.4896739604884348, | |
| "learning_rate": 5.086540930690276e-05, | |
| "loss": 0.227, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.668523676880223, | |
| "grad_norm": 0.471584728228977, | |
| "learning_rate": 4.8701967064566095e-05, | |
| "loss": 0.2199, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.7056638811513465, | |
| "grad_norm": 0.6316398824394676, | |
| "learning_rate": 4.6540955035825676e-05, | |
| "loss": 0.2155, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.7428040854224698, | |
| "grad_norm": 0.5306507117386736, | |
| "learning_rate": 4.438641912789277e-05, | |
| "loss": 0.2216, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.7799442896935933, | |
| "grad_norm": 0.5074632121656725, | |
| "learning_rate": 4.2242393123203986e-05, | |
| "loss": 0.2157, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.8170844939647168, | |
| "grad_norm": 0.6786392166404536, | |
| "learning_rate": 4.011289112726085e-05, | |
| "loss": 0.2121, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.8542246982358404, | |
| "grad_norm": 0.37737812515921004, | |
| "learning_rate": 3.8001900053309184e-05, | |
| "loss": 0.2053, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.8913649025069637, | |
| "grad_norm": 0.48751413968023255, | |
| "learning_rate": 3.591337215792852e-05, | |
| "loss": 0.239, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.9285051067780872, | |
| "grad_norm": 0.6383741405331227, | |
| "learning_rate": 3.3851217641506656e-05, | |
| "loss": 0.2329, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.9656453110492107, | |
| "grad_norm": 0.510851907275428, | |
| "learning_rate": 3.1819297327453045e-05, | |
| "loss": 0.2129, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.4517371979118407, | |
| "learning_rate": 2.9821415433857174e-05, | |
| "loss": 0.2212, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.0371402042711235, | |
| "grad_norm": 0.6477795533764817, | |
| "learning_rate": 2.786131245112495e-05, | |
| "loss": 0.1835, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.074280408542247, | |
| "grad_norm": 0.679636260658929, | |
| "learning_rate": 2.5942658138927867e-05, | |
| "loss": 0.187, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.1114206128133706, | |
| "grad_norm": 0.4922817129749062, | |
| "learning_rate": 2.406904465557614e-05, | |
| "loss": 0.1727, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.148560817084494, | |
| "grad_norm": 0.568983694160024, | |
| "learning_rate": 2.224397983267951e-05, | |
| "loss": 0.1718, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.1857010213556176, | |
| "grad_norm": 0.47125172130466897, | |
| "learning_rate": 2.0470880607686603e-05, | |
| "loss": 0.1611, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.222841225626741, | |
| "grad_norm": 0.5574999246238416, | |
| "learning_rate": 1.8753066626599086e-05, | |
| "loss": 0.1724, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.222841225626741, | |
| "eval_loss": 0.31889092922210693, | |
| "eval_runtime": 69.2279, | |
| "eval_samples_per_second": 5.085, | |
| "eval_steps_per_second": 0.636, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.2599814298978647, | |
| "grad_norm": 0.43455165935100337, | |
| "learning_rate": 1.7093754028837345e-05, | |
| "loss": 0.1786, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.2971216341689877, | |
| "grad_norm": 0.6604753838181086, | |
| "learning_rate": 1.549604942589441e-05, | |
| "loss": 0.1577, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.3342618384401113, | |
| "grad_norm": 0.5420133852475517, | |
| "learning_rate": 1.3962944085050832e-05, | |
| "loss": 0.1631, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.371402042711235, | |
| "grad_norm": 0.4970087479761992, | |
| "learning_rate": 1.2497308329040475e-05, | |
| "loss": 0.1714, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.4085422469823583, | |
| "grad_norm": 0.5146757449079139, | |
| "learning_rate": 1.1101886162151764e-05, | |
| "loss": 0.1637, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.445682451253482, | |
| "grad_norm": 0.5176923026163144, | |
| "learning_rate": 9.779290132826224e-06, | |
| "loss": 0.1728, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.4828226555246053, | |
| "grad_norm": 0.6169350122017051, | |
| "learning_rate": 8.531996442372048e-06, | |
| "loss": 0.1655, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.519962859795729, | |
| "grad_norm": 0.7057265499202516, | |
| "learning_rate": 7.3623403089507825e-06, | |
| "loss": 0.1538, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.5571030640668524, | |
| "grad_norm": 0.5942398255873979, | |
| "learning_rate": 6.272511595516401e-06, | |
| "loss": 0.1601, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.594243268337976, | |
| "grad_norm": 0.5027857142402828, | |
| "learning_rate": 5.264550709892685e-06, | |
| "loss": 0.1706, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.6313834726090994, | |
| "grad_norm": 0.44395576699553374, | |
| "learning_rate": 4.340344784664535e-06, | |
| "loss": 0.1644, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.668523676880223, | |
| "grad_norm": 0.583338230930087, | |
| "learning_rate": 3.501624144035559e-06, | |
| "loss": 0.1712, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.7056638811513465, | |
| "grad_norm": 0.645584661097912, | |
| "learning_rate": 2.7499590642665774e-06, | |
| "loss": 0.1685, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.7428040854224696, | |
| "grad_norm": 0.6016994024894197, | |
| "learning_rate": 2.0867568337605616e-06, | |
| "loss": 0.1488, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.779944289693593, | |
| "grad_norm": 0.4944654437056398, | |
| "learning_rate": 1.5132591182978106e-06, | |
| "loss": 0.17, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.8170844939647166, | |
| "grad_norm": 0.5030466881427367, | |
| "learning_rate": 1.0305396363545717e-06, | |
| "loss": 0.156, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.85422469823584, | |
| "grad_norm": 0.6234438565731387, | |
| "learning_rate": 6.395021488572128e-07, | |
| "loss": 0.1625, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.8913649025069637, | |
| "grad_norm": 0.4826847791868803, | |
| "learning_rate": 3.408787671357494e-07, | |
| "loss": 0.1714, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.928505106778087, | |
| "grad_norm": 0.6381886732044015, | |
| "learning_rate": 1.352285822445065e-07, | |
| "loss": 0.1688, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.9656453110492107, | |
| "grad_norm": 0.5324016878778625, | |
| "learning_rate": 2.2936618216201632e-08, | |
| "loss": 0.1525, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.9656453110492107, | |
| "eval_loss": 0.31994694471359253, | |
| "eval_runtime": 69.2114, | |
| "eval_samples_per_second": 5.086, | |
| "eval_steps_per_second": 0.636, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.9916434540389973, | |
| "step": 807, | |
| "total_flos": 4.48907289034752e+16, | |
| "train_loss": 0.24671251490213614, | |
| "train_runtime": 34121.0002, | |
| "train_samples_per_second": 1.515, | |
| "train_steps_per_second": 0.024 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 807, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.48907289034752e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |