| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.1545595054095827, |
| "eval_steps": 500, |
| "global_step": 500, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.0030911901081916537, |
| "grad_norm": 3.058769941329956, |
| "learning_rate": 1.9972179289026277e-05, |
| "loss": 5.0209, |
| "mean_token_accuracy": 0.21833103336393833, |
| "num_tokens": 28941.0, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.0061823802163833074, |
| "grad_norm": 2.485980749130249, |
| "learning_rate": 1.994126738794436e-05, |
| "loss": 4.7631, |
| "mean_token_accuracy": 0.23807235918939113, |
| "num_tokens": 60990.0, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.00927357032457496, |
| "grad_norm": 5.288544178009033, |
| "learning_rate": 1.9910355486862444e-05, |
| "loss": 4.6799, |
| "mean_token_accuracy": 0.24858475103974342, |
| "num_tokens": 90602.0, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.012364760432766615, |
| "grad_norm": 3.226719379425049, |
| "learning_rate": 1.9879443585780528e-05, |
| "loss": 4.5883, |
| "mean_token_accuracy": 0.2555678006261587, |
| "num_tokens": 121132.0, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.015455950540958269, |
| "grad_norm": 2.027597188949585, |
| "learning_rate": 1.984853168469861e-05, |
| "loss": 4.5125, |
| "mean_token_accuracy": 0.26637452803552153, |
| "num_tokens": 154244.0, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.01854714064914992, |
| "grad_norm": 1.6715962886810303, |
| "learning_rate": 1.9817619783616695e-05, |
| "loss": 4.4224, |
| "mean_token_accuracy": 0.27198897041380404, |
| "num_tokens": 185618.0, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.021638330757341576, |
| "grad_norm": 3.0472540855407715, |
| "learning_rate": 1.9786707882534775e-05, |
| "loss": 4.3464, |
| "mean_token_accuracy": 0.2801041007041931, |
| "num_tokens": 220138.0, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.02472952086553323, |
| "grad_norm": 2.74045729637146, |
| "learning_rate": 1.9755795981452862e-05, |
| "loss": 4.4569, |
| "mean_token_accuracy": 0.2733523309230804, |
| "num_tokens": 252919.0, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.027820710973724884, |
| "grad_norm": 1.9274535179138184, |
| "learning_rate": 1.9724884080370946e-05, |
| "loss": 4.2569, |
| "mean_token_accuracy": 0.2858310595154762, |
| "num_tokens": 284739.0, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.030911901081916538, |
| "grad_norm": 1.906964898109436, |
| "learning_rate": 1.9693972179289026e-05, |
| "loss": 4.2144, |
| "mean_token_accuracy": 0.2953581381589174, |
| "num_tokens": 315186.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.03400309119010819, |
| "grad_norm": 2.042823314666748, |
| "learning_rate": 1.966306027820711e-05, |
| "loss": 4.2595, |
| "mean_token_accuracy": 0.287172843888402, |
| "num_tokens": 347507.0, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.03709428129829984, |
| "grad_norm": 1.9889057874679565, |
| "learning_rate": 1.9632148377125197e-05, |
| "loss": 4.1643, |
| "mean_token_accuracy": 0.30306958928704264, |
| "num_tokens": 380331.0, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.0401854714064915, |
| "grad_norm": 1.7952933311462402, |
| "learning_rate": 1.9601236476043277e-05, |
| "loss": 4.1334, |
| "mean_token_accuracy": 0.3080880597233772, |
| "num_tokens": 410947.0, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.04327666151468315, |
| "grad_norm": 3.8508658409118652, |
| "learning_rate": 1.957032457496136e-05, |
| "loss": 4.0262, |
| "mean_token_accuracy": 0.3118089348077774, |
| "num_tokens": 444130.0, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.04636785162287481, |
| "grad_norm": 31.280546188354492, |
| "learning_rate": 1.9539412673879444e-05, |
| "loss": 3.9889, |
| "mean_token_accuracy": 0.32304659858345985, |
| "num_tokens": 474570.0, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.04945904173106646, |
| "grad_norm": 2.1569416522979736, |
| "learning_rate": 1.950850077279753e-05, |
| "loss": 4.1053, |
| "mean_token_accuracy": 0.31258094161748884, |
| "num_tokens": 505047.0, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.05255023183925812, |
| "grad_norm": 1.808377981185913, |
| "learning_rate": 1.947758887171561e-05, |
| "loss": 3.9711, |
| "mean_token_accuracy": 0.33312112018465995, |
| "num_tokens": 533187.0, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.05564142194744977, |
| "grad_norm": 2.8038811683654785, |
| "learning_rate": 1.9446676970633695e-05, |
| "loss": 3.961, |
| "mean_token_accuracy": 0.33416116759181025, |
| "num_tokens": 565598.0, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.05873261205564142, |
| "grad_norm": 8.930831909179688, |
| "learning_rate": 1.941576506955178e-05, |
| "loss": 4.0107, |
| "mean_token_accuracy": 0.3261258576065302, |
| "num_tokens": 598505.0, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.061823802163833076, |
| "grad_norm": 4.706038475036621, |
| "learning_rate": 1.9384853168469862e-05, |
| "loss": 3.9231, |
| "mean_token_accuracy": 0.33582728281617164, |
| "num_tokens": 630284.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.06491499227202473, |
| "grad_norm": 2.9916040897369385, |
| "learning_rate": 1.9353941267387946e-05, |
| "loss": 3.9407, |
| "mean_token_accuracy": 0.32844844460487366, |
| "num_tokens": 660513.0, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.06800618238021638, |
| "grad_norm": 2.763737678527832, |
| "learning_rate": 1.932302936630603e-05, |
| "loss": 4.0479, |
| "mean_token_accuracy": 0.3248734712600708, |
| "num_tokens": 693154.0, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.07109737248840804, |
| "grad_norm": 3.656487464904785, |
| "learning_rate": 1.9292117465224113e-05, |
| "loss": 3.7843, |
| "mean_token_accuracy": 0.34686593189835546, |
| "num_tokens": 724686.0, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.07418856259659969, |
| "grad_norm": 2.6140244007110596, |
| "learning_rate": 1.9261205564142196e-05, |
| "loss": 3.9261, |
| "mean_token_accuracy": 0.3310456670820713, |
| "num_tokens": 755625.0, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.07727975270479134, |
| "grad_norm": 1.595627784729004, |
| "learning_rate": 1.923029366306028e-05, |
| "loss": 3.841, |
| "mean_token_accuracy": 0.34514380544424056, |
| "num_tokens": 785478.0, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.080370942812983, |
| "grad_norm": 2.0485758781433105, |
| "learning_rate": 1.9199381761978363e-05, |
| "loss": 3.7596, |
| "mean_token_accuracy": 0.35417362824082377, |
| "num_tokens": 816351.0, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.08346213292117466, |
| "grad_norm": 1.7564281225204468, |
| "learning_rate": 1.9168469860896447e-05, |
| "loss": 3.7927, |
| "mean_token_accuracy": 0.3487237967550755, |
| "num_tokens": 849774.0, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.0865533230293663, |
| "grad_norm": 1.5662060976028442, |
| "learning_rate": 1.913755795981453e-05, |
| "loss": 3.7821, |
| "mean_token_accuracy": 0.3511029303073883, |
| "num_tokens": 881474.0, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.08964451313755796, |
| "grad_norm": 2.539433002471924, |
| "learning_rate": 1.9106646058732614e-05, |
| "loss": 3.8613, |
| "mean_token_accuracy": 0.3471171148121357, |
| "num_tokens": 913459.0, |
| "step": 290 |
| }, |
| { |
| "epoch": 0.09273570324574962, |
| "grad_norm": 4.281046390533447, |
| "learning_rate": 1.9075734157650694e-05, |
| "loss": 3.8436, |
| "mean_token_accuracy": 0.3475985363125801, |
| "num_tokens": 947091.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.09582689335394126, |
| "grad_norm": 2.379791736602783, |
| "learning_rate": 1.904482225656878e-05, |
| "loss": 3.8309, |
| "mean_token_accuracy": 0.34195478409528735, |
| "num_tokens": 983471.0, |
| "step": 310 |
| }, |
| { |
| "epoch": 0.09891808346213292, |
| "grad_norm": 2.4176697731018066, |
| "learning_rate": 1.9013910355486865e-05, |
| "loss": 3.75, |
| "mean_token_accuracy": 0.3633588753640652, |
| "num_tokens": 1011575.0, |
| "step": 320 |
| }, |
| { |
| "epoch": 0.10200927357032458, |
| "grad_norm": 3.375523328781128, |
| "learning_rate": 1.898299845440495e-05, |
| "loss": 3.7556, |
| "mean_token_accuracy": 0.36073010191321375, |
| "num_tokens": 1042031.0, |
| "step": 330 |
| }, |
| { |
| "epoch": 0.10510046367851623, |
| "grad_norm": 5.099122524261475, |
| "learning_rate": 1.895208655332303e-05, |
| "loss": 3.6861, |
| "mean_token_accuracy": 0.36301063373684883, |
| "num_tokens": 1074689.0, |
| "step": 340 |
| }, |
| { |
| "epoch": 0.10819165378670788, |
| "grad_norm": 1.4775930643081665, |
| "learning_rate": 1.8921174652241116e-05, |
| "loss": 3.6744, |
| "mean_token_accuracy": 0.3681299857795238, |
| "num_tokens": 1105719.0, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.11128284389489954, |
| "grad_norm": 3.931447744369507, |
| "learning_rate": 1.88902627511592e-05, |
| "loss": 3.6248, |
| "mean_token_accuracy": 0.37062914595007895, |
| "num_tokens": 1136185.0, |
| "step": 360 |
| }, |
| { |
| "epoch": 0.1143740340030912, |
| "grad_norm": 2.6153130531311035, |
| "learning_rate": 1.885935085007728e-05, |
| "loss": 3.6971, |
| "mean_token_accuracy": 0.35686987787485125, |
| "num_tokens": 1166461.0, |
| "step": 370 |
| }, |
| { |
| "epoch": 0.11746522411128284, |
| "grad_norm": 3.0943849086761475, |
| "learning_rate": 1.8828438948995363e-05, |
| "loss": 3.7057, |
| "mean_token_accuracy": 0.36670113652944564, |
| "num_tokens": 1194661.0, |
| "step": 380 |
| }, |
| { |
| "epoch": 0.1205564142194745, |
| "grad_norm": 1.760920524597168, |
| "learning_rate": 1.879752704791345e-05, |
| "loss": 3.6806, |
| "mean_token_accuracy": 0.366318603605032, |
| "num_tokens": 1224670.0, |
| "step": 390 |
| }, |
| { |
| "epoch": 0.12364760432766615, |
| "grad_norm": 1.8976062536239624, |
| "learning_rate": 1.8766615146831534e-05, |
| "loss": 3.7318, |
| "mean_token_accuracy": 0.36500929966568946, |
| "num_tokens": 1252830.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.1267387944358578, |
| "grad_norm": 1.950358510017395, |
| "learning_rate": 1.8735703245749614e-05, |
| "loss": 3.7106, |
| "mean_token_accuracy": 0.3675060346722603, |
| "num_tokens": 1285546.0, |
| "step": 410 |
| }, |
| { |
| "epoch": 0.12982998454404945, |
| "grad_norm": 2.707167148590088, |
| "learning_rate": 1.8704791344667697e-05, |
| "loss": 3.6688, |
| "mean_token_accuracy": 0.37156677842140196, |
| "num_tokens": 1316466.0, |
| "step": 420 |
| }, |
| { |
| "epoch": 0.13292117465224113, |
| "grad_norm": 2.084510564804077, |
| "learning_rate": 1.8673879443585784e-05, |
| "loss": 3.6758, |
| "mean_token_accuracy": 0.3684880450367928, |
| "num_tokens": 1349121.0, |
| "step": 430 |
| }, |
| { |
| "epoch": 0.13601236476043277, |
| "grad_norm": 2.2626636028289795, |
| "learning_rate": 1.8642967542503865e-05, |
| "loss": 3.7214, |
| "mean_token_accuracy": 0.35720510333776473, |
| "num_tokens": 1384366.0, |
| "step": 440 |
| }, |
| { |
| "epoch": 0.1391035548686244, |
| "grad_norm": 2.4145290851593018, |
| "learning_rate": 1.8612055641421948e-05, |
| "loss": 3.7261, |
| "mean_token_accuracy": 0.36429562568664553, |
| "num_tokens": 1415357.0, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.14219474497681608, |
| "grad_norm": 2.7409212589263916, |
| "learning_rate": 1.8581143740340032e-05, |
| "loss": 3.7368, |
| "mean_token_accuracy": 0.3619597226381302, |
| "num_tokens": 1445641.0, |
| "step": 460 |
| }, |
| { |
| "epoch": 0.14528593508500773, |
| "grad_norm": 4.5937275886535645, |
| "learning_rate": 1.8550231839258115e-05, |
| "loss": 3.6911, |
| "mean_token_accuracy": 0.3683665543794632, |
| "num_tokens": 1475360.0, |
| "step": 470 |
| }, |
| { |
| "epoch": 0.14837712519319937, |
| "grad_norm": 12.00837516784668, |
| "learning_rate": 1.85193199381762e-05, |
| "loss": 3.6899, |
| "mean_token_accuracy": 0.3648961283266544, |
| "num_tokens": 1506048.0, |
| "step": 480 |
| }, |
| { |
| "epoch": 0.15146831530139104, |
| "grad_norm": 2.5790181159973145, |
| "learning_rate": 1.8488408037094283e-05, |
| "loss": 3.6948, |
| "mean_token_accuracy": 0.36360194012522695, |
| "num_tokens": 1539998.0, |
| "step": 490 |
| }, |
| { |
| "epoch": 0.1545595054095827, |
| "grad_norm": 1.8515182733535767, |
| "learning_rate": 1.8457496136012366e-05, |
| "loss": 3.65, |
| "mean_token_accuracy": 0.36655392646789553, |
| "num_tokens": 1573641.0, |
| "step": 500 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 6470, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 2, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 906803839715328.0, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
