| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.7991475759190197, | |
| "eval_steps": 20.0, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.015982951518380393, | |
| "grad_norm": 584.8897461802409, | |
| "learning_rate": 3.194888178913738e-07, | |
| "loss": 2.787, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.031965903036760786, | |
| "grad_norm": 36.67515312992242, | |
| "learning_rate": 6.389776357827476e-07, | |
| "loss": 2.5634, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.04794885455514118, | |
| "grad_norm": 10.762606870071352, | |
| "learning_rate": 9.584664536741215e-07, | |
| "loss": 2.0774, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06393180607352157, | |
| "grad_norm": 8.222655227801516, | |
| "learning_rate": 1.2779552715654952e-06, | |
| "loss": 1.7751, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.07991475759190197, | |
| "grad_norm": 9.439874512071311, | |
| "learning_rate": 1.5974440894568691e-06, | |
| "loss": 1.6774, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.09589770911028236, | |
| "grad_norm": 7.064833162397198, | |
| "learning_rate": 1.916932907348243e-06, | |
| "loss": 1.5861, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.11188066062866275, | |
| "grad_norm": 10.311171109232708, | |
| "learning_rate": 2.2364217252396165e-06, | |
| "loss": 1.5787, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.12786361214704314, | |
| "grad_norm": 9.640566718774098, | |
| "learning_rate": 2.5559105431309904e-06, | |
| "loss": 1.5406, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.14384656366542356, | |
| "grad_norm": 7.158156117158717, | |
| "learning_rate": 2.8753993610223648e-06, | |
| "loss": 1.5008, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.15982951518380395, | |
| "grad_norm": 6.670618524707829, | |
| "learning_rate": 3.1948881789137383e-06, | |
| "loss": 1.4853, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.17581246670218434, | |
| "grad_norm": 6.9280590132917075, | |
| "learning_rate": 3.514376996805112e-06, | |
| "loss": 1.4688, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.19179541822056473, | |
| "grad_norm": 6.984532799464284, | |
| "learning_rate": 3.833865814696486e-06, | |
| "loss": 1.4501, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.20777836973894512, | |
| "grad_norm": 4.6035696826309165, | |
| "learning_rate": 4.15335463258786e-06, | |
| "loss": 1.4325, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.2237613212573255, | |
| "grad_norm": 7.6475074547231054, | |
| "learning_rate": 4.472843450479233e-06, | |
| "loss": 1.4058, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.23974427277570592, | |
| "grad_norm": 5.215774162230619, | |
| "learning_rate": 4.792332268370608e-06, | |
| "loss": 1.3873, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.2557272242940863, | |
| "grad_norm": 5.284228741443329, | |
| "learning_rate": 5.111821086261981e-06, | |
| "loss": 1.3882, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.2717101758124667, | |
| "grad_norm": 4.315985648750639, | |
| "learning_rate": 5.431309904153355e-06, | |
| "loss": 1.3684, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.2876931273308471, | |
| "grad_norm": 4.037073137808611, | |
| "learning_rate": 5.7507987220447296e-06, | |
| "loss": 1.359, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.3036760788492275, | |
| "grad_norm": 6.346300332686573, | |
| "learning_rate": 6.070287539936103e-06, | |
| "loss": 1.3435, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.3196590303676079, | |
| "grad_norm": 4.63092372720661, | |
| "learning_rate": 6.3897763578274765e-06, | |
| "loss": 1.3468, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.33564198188598826, | |
| "grad_norm": 5.057903719523363, | |
| "learning_rate": 6.709265175718851e-06, | |
| "loss": 1.3341, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.3516249334043687, | |
| "grad_norm": 5.25790384573804, | |
| "learning_rate": 7.028753993610224e-06, | |
| "loss": 1.3233, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.3676078849227491, | |
| "grad_norm": 3.3267127292309517, | |
| "learning_rate": 7.348242811501598e-06, | |
| "loss": 1.3025, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.38359083644112946, | |
| "grad_norm": 4.174952594356688, | |
| "learning_rate": 7.667731629392972e-06, | |
| "loss": 1.3, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.3995737879595099, | |
| "grad_norm": 2.7547093096631166, | |
| "learning_rate": 7.987220447284347e-06, | |
| "loss": 1.299, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.41555673947789024, | |
| "grad_norm": 3.0076000117279023, | |
| "learning_rate": 8.30670926517572e-06, | |
| "loss": 1.2789, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.43153969099627065, | |
| "grad_norm": 3.6118960867167695, | |
| "learning_rate": 8.626198083067093e-06, | |
| "loss": 1.289, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.447522642514651, | |
| "grad_norm": 4.187154082722337, | |
| "learning_rate": 8.945686900958466e-06, | |
| "loss": 1.2772, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.46350559403303143, | |
| "grad_norm": 3.3209936896861265, | |
| "learning_rate": 9.265175718849841e-06, | |
| "loss": 1.2714, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.47948854555141185, | |
| "grad_norm": 3.4443001562081874, | |
| "learning_rate": 9.584664536741216e-06, | |
| "loss": 1.2615, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4954714970697922, | |
| "grad_norm": 3.4861656497713764, | |
| "learning_rate": 9.904153354632589e-06, | |
| "loss": 1.2756, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.5114544485881726, | |
| "grad_norm": 3.2211114109625165, | |
| "learning_rate": 9.999847101583393e-06, | |
| "loss": 1.2434, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.527437400106553, | |
| "grad_norm": 3.1006035508008485, | |
| "learning_rate": 9.999098233890869e-06, | |
| "loss": 1.2489, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.5434203516249334, | |
| "grad_norm": 3.0983188498194516, | |
| "learning_rate": 9.997725406892392e-06, | |
| "loss": 1.2407, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.5594033031433138, | |
| "grad_norm": 3.350092298430153, | |
| "learning_rate": 9.995728791936505e-06, | |
| "loss": 1.243, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.5753862546616942, | |
| "grad_norm": 2.712885657555514, | |
| "learning_rate": 9.993108638229449e-06, | |
| "loss": 1.2677, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.5913692061800746, | |
| "grad_norm": 2.6927845803955983, | |
| "learning_rate": 9.989865272804064e-06, | |
| "loss": 1.2272, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.607352157698455, | |
| "grad_norm": 2.9181755886200675, | |
| "learning_rate": 9.985999100478964e-06, | |
| "loss": 1.2217, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.6233351092168353, | |
| "grad_norm": 2.663582794855719, | |
| "learning_rate": 9.981510603808024e-06, | |
| "loss": 1.2225, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.6393180607352158, | |
| "grad_norm": 2.81821319928569, | |
| "learning_rate": 9.976400343020134e-06, | |
| "loss": 1.2291, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6553010122535962, | |
| "grad_norm": 2.7190463266479363, | |
| "learning_rate": 9.970668955949285e-06, | |
| "loss": 1.2017, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.6712839637719765, | |
| "grad_norm": 2.6121849551347056, | |
| "learning_rate": 9.964317157954955e-06, | |
| "loss": 1.1999, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.687266915290357, | |
| "grad_norm": 2.8153860907477988, | |
| "learning_rate": 9.95734574183282e-06, | |
| "loss": 1.2119, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.7032498668087374, | |
| "grad_norm": 2.8604210593081016, | |
| "learning_rate": 9.949755577715806e-06, | |
| "loss": 1.1973, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.7192328183271177, | |
| "grad_norm": 2.683531260179304, | |
| "learning_rate": 9.941547612965475e-06, | |
| "loss": 1.2138, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.7352157698454982, | |
| "grad_norm": 3.0411589457571377, | |
| "learning_rate": 9.932722872053797e-06, | |
| "loss": 1.2221, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.7511987213638786, | |
| "grad_norm": 2.993913433024178, | |
| "learning_rate": 9.923282456435262e-06, | |
| "loss": 1.209, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.7671816728822589, | |
| "grad_norm": 2.6707946625116072, | |
| "learning_rate": 9.913227544409416e-06, | |
| "loss": 1.1918, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.7831646244006393, | |
| "grad_norm": 2.7683148756228033, | |
| "learning_rate": 9.90255939097379e-06, | |
| "loss": 1.1784, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.7991475759190197, | |
| "grad_norm": 2.803374407208519, | |
| "learning_rate": 9.891279327667252e-06, | |
| "loss": 1.1769, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 3125, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 100, | |
| "total_flos": 7.826917846385951e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |