{ "best_metric": null, "best_model_checkpoint": null, "epoch": 12.269938650306749, "eval_steps": 10000000, "global_step": 1000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.12269938650306748, "grad_norm": 30.28219704365505, "learning_rate": 2.4691358024691355e-08, "loss": 2.9799, "step": 10 }, { "epoch": 0.24539877300613497, "grad_norm": 30.66075380374518, "learning_rate": 4.938271604938271e-08, "loss": 2.9842, "step": 20 }, { "epoch": 0.36809815950920244, "grad_norm": 29.626899456192994, "learning_rate": 7.407407407407407e-08, "loss": 3.015, "step": 30 }, { "epoch": 0.49079754601226994, "grad_norm": 29.470505364249597, "learning_rate": 9.876543209876542e-08, "loss": 2.9547, "step": 40 }, { "epoch": 0.6134969325153374, "grad_norm": 27.407219480683995, "learning_rate": 1.2345679012345677e-07, "loss": 2.8784, "step": 50 }, { "epoch": 0.7361963190184049, "grad_norm": 20.68736366691469, "learning_rate": 1.4814814814814815e-07, "loss": 2.7798, "step": 60 }, { "epoch": 0.8588957055214724, "grad_norm": 17.11747188448127, "learning_rate": 1.728395061728395e-07, "loss": 2.5723, "step": 70 }, { "epoch": 0.9815950920245399, "grad_norm": 7.782003292603646, "learning_rate": 1.9753086419753084e-07, "loss": 2.4467, "step": 80 }, { "epoch": 1.1042944785276074, "grad_norm": 4.365497818859387, "learning_rate": 2.222222222222222e-07, "loss": 2.2687, "step": 90 }, { "epoch": 1.2269938650306749, "grad_norm": 3.3711421712732283, "learning_rate": 2.4691358024691354e-07, "loss": 2.1767, "step": 100 }, { "epoch": 1.3496932515337423, "grad_norm": 2.956591715084401, "learning_rate": 2.716049382716049e-07, "loss": 2.1332, "step": 110 }, { "epoch": 1.4723926380368098, "grad_norm": 2.635796811630019, "learning_rate": 2.962962962962963e-07, "loss": 2.1135, "step": 120 }, { "epoch": 1.5950920245398774, "grad_norm": 2.3273823610234876, "learning_rate": 3.209876543209876e-07, "loss": 2.1097, "step": 130 }, { "epoch": 1.7177914110429446, "grad_norm": 2.2617900128343833, "learning_rate": 3.45679012345679e-07, "loss": 2.1161, "step": 140 }, { "epoch": 1.8404907975460123, "grad_norm": 2.1638223937937076, "learning_rate": 3.703703703703703e-07, "loss": 2.0616, "step": 150 }, { "epoch": 1.9631901840490797, "grad_norm": 2.1437544811484854, "learning_rate": 3.950617283950617e-07, "loss": 2.0815, "step": 160 }, { "epoch": 2.085889570552147, "grad_norm": 2.1420606825819197, "learning_rate": 4.1975308641975306e-07, "loss": 2.063, "step": 170 }, { "epoch": 2.208588957055215, "grad_norm": 2.144143735036332, "learning_rate": 4.444444444444444e-07, "loss": 2.0513, "step": 180 }, { "epoch": 2.331288343558282, "grad_norm": 2.09031994519009, "learning_rate": 4.6913580246913576e-07, "loss": 2.0583, "step": 190 }, { "epoch": 2.4539877300613497, "grad_norm": 2.074953413736776, "learning_rate": 4.938271604938271e-07, "loss": 2.0446, "step": 200 }, { "epoch": 2.5766871165644174, "grad_norm": 2.0364457610758473, "learning_rate": 5.185185185185185e-07, "loss": 2.0336, "step": 210 }, { "epoch": 2.6993865030674846, "grad_norm": 2.09662270077011, "learning_rate": 5.432098765432098e-07, "loss": 2.03, "step": 220 }, { "epoch": 2.8220858895705523, "grad_norm": 2.0184952150772273, "learning_rate": 5.679012345679012e-07, "loss": 2.0197, "step": 230 }, { "epoch": 2.9447852760736195, "grad_norm": 2.2026915181725535, "learning_rate": 5.925925925925926e-07, "loss": 2.007, "step": 240 }, { "epoch": 3.067484662576687, "grad_norm": 1.9452709028668997, "learning_rate": 6.172839506172839e-07, "loss": 2.0014, "step": 250 }, { "epoch": 3.190184049079755, "grad_norm": 1.933109075200724, "learning_rate": 6.419753086419752e-07, "loss": 1.977, "step": 260 }, { "epoch": 3.312883435582822, "grad_norm": 1.9739789693701668, "learning_rate": 6.666666666666666e-07, "loss": 1.9816, "step": 270 }, { "epoch": 3.4355828220858897, "grad_norm": 2.0255729355536625, "learning_rate": 6.91358024691358e-07, "loss": 2.0138, "step": 280 }, { "epoch": 3.558282208588957, "grad_norm": 2.109260451722745, "learning_rate": 7.160493827160494e-07, "loss": 1.9852, "step": 290 }, { "epoch": 3.6809815950920246, "grad_norm": 2.02230955779347, "learning_rate": 7.407407407407406e-07, "loss": 1.989, "step": 300 }, { "epoch": 3.8036809815950923, "grad_norm": 2.114117608740877, "learning_rate": 7.65432098765432e-07, "loss": 1.9656, "step": 310 }, { "epoch": 3.9263803680981595, "grad_norm": 1.9608507920028475, "learning_rate": 7.901234567901234e-07, "loss": 1.9819, "step": 320 }, { "epoch": 4.049079754601227, "grad_norm": 1.992359665195648, "learning_rate": 8.148148148148147e-07, "loss": 1.9517, "step": 330 }, { "epoch": 4.171779141104294, "grad_norm": 2.0184004724781524, "learning_rate": 8.395061728395061e-07, "loss": 1.8995, "step": 340 }, { "epoch": 4.294478527607362, "grad_norm": 1.9438721415658846, "learning_rate": 8.641975308641974e-07, "loss": 1.9493, "step": 350 }, { "epoch": 4.41717791411043, "grad_norm": 2.037329633517169, "learning_rate": 8.888888888888888e-07, "loss": 1.9592, "step": 360 }, { "epoch": 4.539877300613497, "grad_norm": 2.045589243795842, "learning_rate": 9.135802469135801e-07, "loss": 1.9269, "step": 370 }, { "epoch": 4.662576687116564, "grad_norm": 2.0376567384336814, "learning_rate": 9.382716049382715e-07, "loss": 1.9091, "step": 380 }, { "epoch": 4.785276073619632, "grad_norm": 2.151886931601921, "learning_rate": 9.629629629629628e-07, "loss": 1.9338, "step": 390 }, { "epoch": 4.9079754601226995, "grad_norm": 2.1016930186631955, "learning_rate": 9.876543209876542e-07, "loss": 1.9361, "step": 400 }, { "epoch": 5.030674846625767, "grad_norm": 1.8425197677510754, "learning_rate": 9.999953571567085e-07, "loss": 1.8946, "step": 410 }, { "epoch": 5.153374233128835, "grad_norm": 2.1108335201344826, "learning_rate": 9.999582149277185e-07, "loss": 1.8636, "step": 420 }, { "epoch": 5.276073619631902, "grad_norm": 1.965491417959117, "learning_rate": 9.99883933228855e-07, "loss": 1.8863, "step": 430 }, { "epoch": 5.398773006134969, "grad_norm": 2.0748380468153442, "learning_rate": 9.997725175781443e-07, "loss": 1.8708, "step": 440 }, { "epoch": 5.521472392638037, "grad_norm": 2.0061874641389994, "learning_rate": 9.99623976252115e-07, "loss": 1.8629, "step": 450 }, { "epoch": 5.644171779141105, "grad_norm": 1.9227878063794117, "learning_rate": 9.994383202851812e-07, "loss": 1.8574, "step": 460 }, { "epoch": 5.766871165644172, "grad_norm": 2.022882706029264, "learning_rate": 9.992155634688238e-07, "loss": 1.8489, "step": 470 }, { "epoch": 5.889570552147239, "grad_norm": 1.9151165000656245, "learning_rate": 9.98955722350566e-07, "loss": 1.8702, "step": 480 }, { "epoch": 6.012269938650307, "grad_norm": 1.8932452511409288, "learning_rate": 9.986588162327434e-07, "loss": 1.8569, "step": 490 }, { "epoch": 6.134969325153374, "grad_norm": 2.0203097592372035, "learning_rate": 9.983248671710714e-07, "loss": 1.7873, "step": 500 }, { "epoch": 6.257668711656442, "grad_norm": 1.9081758147724377, "learning_rate": 9.979538999730047e-07, "loss": 1.779, "step": 510 }, { "epoch": 6.38036809815951, "grad_norm": 2.072946207781536, "learning_rate": 9.975459421958967e-07, "loss": 1.8006, "step": 520 }, { "epoch": 6.5030674846625764, "grad_norm": 2.0035469674735893, "learning_rate": 9.971010241449513e-07, "loss": 1.8018, "step": 530 }, { "epoch": 6.625766871165644, "grad_norm": 2.0835946373439143, "learning_rate": 9.966191788709714e-07, "loss": 1.8, "step": 540 }, { "epoch": 6.748466257668712, "grad_norm": 2.051477600119681, "learning_rate": 9.961004421679046e-07, "loss": 1.7869, "step": 550 }, { "epoch": 6.871165644171779, "grad_norm": 1.9111605839024262, "learning_rate": 9.955448525701835e-07, "loss": 1.7929, "step": 560 }, { "epoch": 6.993865030674847, "grad_norm": 2.036265101667604, "learning_rate": 9.949524513498636e-07, "loss": 1.8098, "step": 570 }, { "epoch": 7.116564417177914, "grad_norm": 1.9440590850722006, "learning_rate": 9.943232825135566e-07, "loss": 1.7158, "step": 580 }, { "epoch": 7.2392638036809815, "grad_norm": 1.9745053511466075, "learning_rate": 9.93657392799163e-07, "loss": 1.7269, "step": 590 }, { "epoch": 7.361963190184049, "grad_norm": 2.0172262033751522, "learning_rate": 9.92954831672398e-07, "loss": 1.7173, "step": 600 }, { "epoch": 7.484662576687117, "grad_norm": 2.017376130474714, "learning_rate": 9.922156513231197e-07, "loss": 1.723, "step": 610 }, { "epoch": 7.6073619631901845, "grad_norm": 2.129118497262339, "learning_rate": 9.914399066614487e-07, "loss": 1.7109, "step": 620 }, { "epoch": 7.730061349693251, "grad_norm": 2.0003358633590986, "learning_rate": 9.906276553136922e-07, "loss": 1.7287, "step": 630 }, { "epoch": 7.852760736196319, "grad_norm": 2.0380074509433554, "learning_rate": 9.897789576180616e-07, "loss": 1.7128, "step": 640 }, { "epoch": 7.975460122699387, "grad_norm": 1.972210672677589, "learning_rate": 9.888938766201907e-07, "loss": 1.7209, "step": 650 }, { "epoch": 8.098159509202453, "grad_norm": 2.0353896296888263, "learning_rate": 9.879724780684517e-07, "loss": 1.6636, "step": 660 }, { "epoch": 8.220858895705522, "grad_norm": 2.1575258154001067, "learning_rate": 9.87014830409073e-07, "loss": 1.6493, "step": 670 }, { "epoch": 8.343558282208589, "grad_norm": 2.0731699660667156, "learning_rate": 9.860210047810515e-07, "loss": 1.627, "step": 680 }, { "epoch": 8.466257668711656, "grad_norm": 2.1169883616519476, "learning_rate": 9.849910750108717e-07, "loss": 1.6516, "step": 690 }, { "epoch": 8.588957055214724, "grad_norm": 2.0906272175984917, "learning_rate": 9.839251176070183e-07, "loss": 1.6619, "step": 700 }, { "epoch": 8.71165644171779, "grad_norm": 2.059390905577669, "learning_rate": 9.828232117542947e-07, "loss": 1.6457, "step": 710 }, { "epoch": 8.83435582822086, "grad_norm": 2.111680195842936, "learning_rate": 9.816854393079402e-07, "loss": 1.6178, "step": 720 }, { "epoch": 8.957055214723926, "grad_norm": 2.144260583400968, "learning_rate": 9.805118847875487e-07, "loss": 1.6382, "step": 730 }, { "epoch": 9.079754601226995, "grad_norm": 2.285829860466323, "learning_rate": 9.793026353707914e-07, "loss": 1.5706, "step": 740 }, { "epoch": 9.202453987730062, "grad_norm": 2.153794951665122, "learning_rate": 9.780577808869398e-07, "loss": 1.5595, "step": 750 }, { "epoch": 9.325153374233128, "grad_norm": 2.138270605360391, "learning_rate": 9.767774138101934e-07, "loss": 1.5649, "step": 760 }, { "epoch": 9.447852760736197, "grad_norm": 2.140711587703231, "learning_rate": 9.754616292528093e-07, "loss": 1.5466, "step": 770 }, { "epoch": 9.570552147239264, "grad_norm": 2.221273893753507, "learning_rate": 9.74110524958038e-07, "loss": 1.5284, "step": 780 }, { "epoch": 9.69325153374233, "grad_norm": 2.3103597560152545, "learning_rate": 9.72724201292862e-07, "loss": 1.5592, "step": 790 }, { "epoch": 9.815950920245399, "grad_norm": 2.2066393219378373, "learning_rate": 9.713027612405394e-07, "loss": 1.546, "step": 800 }, { "epoch": 9.938650306748466, "grad_norm": 2.274268167957973, "learning_rate": 9.698463103929541e-07, "loss": 1.5556, "step": 810 }, { "epoch": 10.061349693251534, "grad_norm": 2.4288759947142666, "learning_rate": 9.68354956942773e-07, "loss": 1.5192, "step": 820 }, { "epoch": 10.184049079754601, "grad_norm": 2.56196295256849, "learning_rate": 9.668288116754076e-07, "loss": 1.4731, "step": 830 }, { "epoch": 10.30674846625767, "grad_norm": 2.351647479927511, "learning_rate": 9.652679879607843e-07, "loss": 1.4728, "step": 840 }, { "epoch": 10.429447852760736, "grad_norm": 2.3330020594718897, "learning_rate": 9.636726017449236e-07, "loss": 1.4558, "step": 850 }, { "epoch": 10.552147239263803, "grad_norm": 2.420606264665063, "learning_rate": 9.62042771541326e-07, "loss": 1.4287, "step": 860 }, { "epoch": 10.674846625766872, "grad_norm": 2.3947612477487867, "learning_rate": 9.603786184221692e-07, "loss": 1.4469, "step": 870 }, { "epoch": 10.797546012269938, "grad_norm": 2.3399004198935303, "learning_rate": 9.586802660093136e-07, "loss": 1.4396, "step": 880 }, { "epoch": 10.920245398773005, "grad_norm": 2.387016507283165, "learning_rate": 9.56947840465119e-07, "loss": 1.4595, "step": 890 }, { "epoch": 11.042944785276074, "grad_norm": 2.533848992609839, "learning_rate": 9.551814704830734e-07, "loss": 1.417, "step": 900 }, { "epoch": 11.16564417177914, "grad_norm": 2.5734232080415254, "learning_rate": 9.533812872782313e-07, "loss": 1.348, "step": 910 }, { "epoch": 11.28834355828221, "grad_norm": 2.6920925651578087, "learning_rate": 9.515474245774684e-07, "loss": 1.3556, "step": 920 }, { "epoch": 11.411042944785276, "grad_norm": 2.7013002533947574, "learning_rate": 9.496800186095465e-07, "loss": 1.3539, "step": 930 }, { "epoch": 11.533742331288344, "grad_norm": 2.731007254322984, "learning_rate": 9.477792080949938e-07, "loss": 1.3436, "step": 940 }, { "epoch": 11.656441717791411, "grad_norm": 2.8652743906067375, "learning_rate": 9.458451342358e-07, "loss": 1.3521, "step": 950 }, { "epoch": 11.779141104294478, "grad_norm": 2.5726368131527075, "learning_rate": 9.43877940704928e-07, "loss": 1.3426, "step": 960 }, { "epoch": 11.901840490797547, "grad_norm": 2.6814604487668463, "learning_rate": 9.418777736356393e-07, "loss": 1.3419, "step": 970 }, { "epoch": 12.024539877300613, "grad_norm": 3.1868357701860774, "learning_rate": 9.39844781610641e-07, "loss": 1.3155, "step": 980 }, { "epoch": 12.14723926380368, "grad_norm": 3.082069159033982, "learning_rate": 9.377791156510454e-07, "loss": 1.2428, "step": 990 }, { "epoch": 12.269938650306749, "grad_norm": 3.075730479893534, "learning_rate": 9.356809292051539e-07, "loss": 1.224, "step": 1000 } ], "logging_steps": 10, "max_steps": 4050, "num_input_tokens_seen": 0, "num_train_epochs": 50, "save_steps": 1000, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 92928324206592.0, "train_batch_size": 8, "trial_name": null, "trial_params": null }