| { | |
| "best_global_step": 1400, | |
| "best_metric": 0.6364541053771973, | |
| "best_model_checkpoint": "/scr/dhanda/projects/sampling_inference/trained_models/aswin_llama_star/checkpoint-1400", | |
| "epoch": 1.0, | |
| "eval_steps": 100, | |
| "global_step": 1441, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.006939625260235947, | |
| "grad_norm": 8.436548233032227, | |
| "learning_rate": 2.7586206896551725e-06, | |
| "loss": 0.8313, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.013879250520471894, | |
| "grad_norm": 5.655324935913086, | |
| "learning_rate": 6.206896551724138e-06, | |
| "loss": 0.5995, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.020818875780707843, | |
| "grad_norm": 6.097832202911377, | |
| "learning_rate": 9.655172413793103e-06, | |
| "loss": 0.5897, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.027758501040943788, | |
| "grad_norm": 5.244992256164551, | |
| "learning_rate": 1.310344827586207e-05, | |
| "loss": 0.5975, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03469812630117974, | |
| "grad_norm": 5.82904577255249, | |
| "learning_rate": 1.6551724137931037e-05, | |
| "loss": 0.5866, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.041637751561415685, | |
| "grad_norm": 4.755958557128906, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6143, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.048577376821651634, | |
| "grad_norm": 4.591907024383545, | |
| "learning_rate": 2.3448275862068967e-05, | |
| "loss": 0.6709, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.055517002081887576, | |
| "grad_norm": 5.520663738250732, | |
| "learning_rate": 2.689655172413793e-05, | |
| "loss": 0.6815, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.062456627342123525, | |
| "grad_norm": 5.258326530456543, | |
| "learning_rate": 3.0344827586206897e-05, | |
| "loss": 0.7171, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.06939625260235947, | |
| "grad_norm": 4.56249475479126, | |
| "learning_rate": 3.3793103448275865e-05, | |
| "loss": 0.7289, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.06939625260235947, | |
| "eval_loss": 0.7279375195503235, | |
| "eval_runtime": 49.9788, | |
| "eval_samples_per_second": 28.832, | |
| "eval_steps_per_second": 7.223, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07633587786259542, | |
| "grad_norm": 3.349087715148926, | |
| "learning_rate": 3.724137931034483e-05, | |
| "loss": 0.7553, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.08327550312283137, | |
| "grad_norm": 3.4430437088012695, | |
| "learning_rate": 4.0689655172413795e-05, | |
| "loss": 0.7228, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.09021512838306732, | |
| "grad_norm": 4.375394821166992, | |
| "learning_rate": 4.413793103448276e-05, | |
| "loss": 0.8678, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.09715475364330327, | |
| "grad_norm": 4.496102333068848, | |
| "learning_rate": 4.7586206896551725e-05, | |
| "loss": 0.8478, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.1040943789035392, | |
| "grad_norm": 4.770753383636475, | |
| "learning_rate": 4.988425925925926e-05, | |
| "loss": 0.824, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.11103400416377515, | |
| "grad_norm": 4.819082260131836, | |
| "learning_rate": 4.949845679012346e-05, | |
| "loss": 0.8833, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.1179736294240111, | |
| "grad_norm": 5.851232528686523, | |
| "learning_rate": 4.911265432098766e-05, | |
| "loss": 0.8861, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.12491325468424705, | |
| "grad_norm": 3.9091575145721436, | |
| "learning_rate": 4.8726851851851855e-05, | |
| "loss": 0.9043, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.131852879944483, | |
| "grad_norm": 4.504231929779053, | |
| "learning_rate": 4.834104938271605e-05, | |
| "loss": 0.909, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.13879250520471895, | |
| "grad_norm": 5.178035259246826, | |
| "learning_rate": 4.795524691358025e-05, | |
| "loss": 0.8768, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.13879250520471895, | |
| "eval_loss": 0.8475821018218994, | |
| "eval_runtime": 49.8841, | |
| "eval_samples_per_second": 28.887, | |
| "eval_steps_per_second": 7.237, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.1457321304649549, | |
| "grad_norm": 4.2687907218933105, | |
| "learning_rate": 4.756944444444444e-05, | |
| "loss": 0.7687, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.15267175572519084, | |
| "grad_norm": 3.8715438842773438, | |
| "learning_rate": 4.7183641975308646e-05, | |
| "loss": 0.8323, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.1596113809854268, | |
| "grad_norm": 5.992573261260986, | |
| "learning_rate": 4.679783950617284e-05, | |
| "loss": 0.829, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.16655100624566274, | |
| "grad_norm": 3.7021474838256836, | |
| "learning_rate": 4.6412037037037034e-05, | |
| "loss": 0.8627, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.1734906315058987, | |
| "grad_norm": 3.4208054542541504, | |
| "learning_rate": 4.602623456790124e-05, | |
| "loss": 0.9109, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.18043025676613464, | |
| "grad_norm": 3.720872163772583, | |
| "learning_rate": 4.5640432098765436e-05, | |
| "loss": 0.8255, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.1873698820263706, | |
| "grad_norm": 3.8880999088287354, | |
| "learning_rate": 4.525462962962963e-05, | |
| "loss": 0.8802, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.19430950728660654, | |
| "grad_norm": 3.356327772140503, | |
| "learning_rate": 4.486882716049383e-05, | |
| "loss": 0.9039, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.20124913254684246, | |
| "grad_norm": 3.6388580799102783, | |
| "learning_rate": 4.448302469135803e-05, | |
| "loss": 0.8776, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.2081887578070784, | |
| "grad_norm": 4.57245397567749, | |
| "learning_rate": 4.4097222222222226e-05, | |
| "loss": 0.8568, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.2081887578070784, | |
| "eval_loss": 0.8598970770835876, | |
| "eval_runtime": 50.2014, | |
| "eval_samples_per_second": 28.704, | |
| "eval_steps_per_second": 7.191, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.21512838306731435, | |
| "grad_norm": 3.0655972957611084, | |
| "learning_rate": 4.3711419753086424e-05, | |
| "loss": 0.8248, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.2220680083275503, | |
| "grad_norm": 3.4082236289978027, | |
| "learning_rate": 4.332561728395062e-05, | |
| "loss": 0.8737, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.22900763358778625, | |
| "grad_norm": 4.789612770080566, | |
| "learning_rate": 4.293981481481482e-05, | |
| "loss": 0.8218, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.2359472588480222, | |
| "grad_norm": 3.7866945266723633, | |
| "learning_rate": 4.255401234567901e-05, | |
| "loss": 0.8208, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.24288688410825815, | |
| "grad_norm": 4.31322717666626, | |
| "learning_rate": 4.2168209876543214e-05, | |
| "loss": 0.7825, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.2498265093684941, | |
| "grad_norm": 3.4791786670684814, | |
| "learning_rate": 4.178240740740741e-05, | |
| "loss": 0.8161, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.25676613462873005, | |
| "grad_norm": 4.521151542663574, | |
| "learning_rate": 4.13966049382716e-05, | |
| "loss": 0.7744, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.263705759888966, | |
| "grad_norm": 2.1991310119628906, | |
| "learning_rate": 4.104938271604938e-05, | |
| "loss": 0.8381, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.27064538514920194, | |
| "grad_norm": 3.745811939239502, | |
| "learning_rate": 4.066358024691358e-05, | |
| "loss": 0.9447, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.2775850104094379, | |
| "grad_norm": 3.5998716354370117, | |
| "learning_rate": 4.027777777777778e-05, | |
| "loss": 0.8452, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.2775850104094379, | |
| "eval_loss": 0.8371486067771912, | |
| "eval_runtime": 50.1497, | |
| "eval_samples_per_second": 28.734, | |
| "eval_steps_per_second": 7.198, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.28452463566967384, | |
| "grad_norm": 4.008808612823486, | |
| "learning_rate": 3.9891975308641976e-05, | |
| "loss": 0.8101, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.2914642609299098, | |
| "grad_norm": 4.993364334106445, | |
| "learning_rate": 3.950617283950617e-05, | |
| "loss": 0.9292, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.29840388619014574, | |
| "grad_norm": 3.597254514694214, | |
| "learning_rate": 3.912037037037037e-05, | |
| "loss": 0.8176, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.3053435114503817, | |
| "grad_norm": 4.86491584777832, | |
| "learning_rate": 3.873456790123457e-05, | |
| "loss": 0.7595, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.31228313671061764, | |
| "grad_norm": 4.191522121429443, | |
| "learning_rate": 3.8348765432098766e-05, | |
| "loss": 0.8463, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.3192227619708536, | |
| "grad_norm": 3.8313114643096924, | |
| "learning_rate": 3.7962962962962964e-05, | |
| "loss": 0.815, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.32616238723108953, | |
| "grad_norm": 2.289886951446533, | |
| "learning_rate": 3.757716049382716e-05, | |
| "loss": 0.7598, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.3331020124913255, | |
| "grad_norm": 3.6466236114501953, | |
| "learning_rate": 3.719135802469136e-05, | |
| "loss": 0.7771, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.34004163775156143, | |
| "grad_norm": 3.2481741905212402, | |
| "learning_rate": 3.6805555555555556e-05, | |
| "loss": 0.7833, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.3469812630117974, | |
| "grad_norm": 5.6007208824157715, | |
| "learning_rate": 3.6419753086419754e-05, | |
| "loss": 0.8243, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.3469812630117974, | |
| "eval_loss": 0.8030129671096802, | |
| "eval_runtime": 50.3084, | |
| "eval_samples_per_second": 28.643, | |
| "eval_steps_per_second": 7.176, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.35392088827203333, | |
| "grad_norm": 4.155673503875732, | |
| "learning_rate": 3.603395061728395e-05, | |
| "loss": 0.7968, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.3608605135322693, | |
| "grad_norm": 3.114962577819824, | |
| "learning_rate": 3.564814814814815e-05, | |
| "loss": 0.815, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.3678001387925052, | |
| "grad_norm": 4.049103736877441, | |
| "learning_rate": 3.526234567901235e-05, | |
| "loss": 0.8768, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.3747397640527412, | |
| "grad_norm": 3.64631724357605, | |
| "learning_rate": 3.4876543209876545e-05, | |
| "loss": 0.7909, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.3816793893129771, | |
| "grad_norm": 3.093494176864624, | |
| "learning_rate": 3.449074074074074e-05, | |
| "loss": 0.7576, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.3886190145732131, | |
| "grad_norm": 3.2251219749450684, | |
| "learning_rate": 3.410493827160494e-05, | |
| "loss": 0.813, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.39555863983344897, | |
| "grad_norm": 2.702892303466797, | |
| "learning_rate": 3.371913580246914e-05, | |
| "loss": 0.7765, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.4024982650936849, | |
| "grad_norm": 3.3482439517974854, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.8653, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.40943789035392086, | |
| "grad_norm": 2.537188768386841, | |
| "learning_rate": 3.294753086419753e-05, | |
| "loss": 0.7409, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.4163775156141568, | |
| "grad_norm": 3.178288459777832, | |
| "learning_rate": 3.256172839506173e-05, | |
| "loss": 0.7191, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.4163775156141568, | |
| "eval_loss": 0.7933737635612488, | |
| "eval_runtime": 50.9506, | |
| "eval_samples_per_second": 28.282, | |
| "eval_steps_per_second": 7.085, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.42331714087439276, | |
| "grad_norm": 3.4489400386810303, | |
| "learning_rate": 3.221450617283951e-05, | |
| "loss": 0.8456, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.4302567661346287, | |
| "grad_norm": 3.6795783042907715, | |
| "learning_rate": 3.182870370370371e-05, | |
| "loss": 0.8392, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.43719639139486466, | |
| "grad_norm": 2.7610459327697754, | |
| "learning_rate": 3.14429012345679e-05, | |
| "loss": 0.7238, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.4441360166551006, | |
| "grad_norm": 3.525322675704956, | |
| "learning_rate": 3.10570987654321e-05, | |
| "loss": 0.8437, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.45107564191533656, | |
| "grad_norm": 4.020907402038574, | |
| "learning_rate": 3.06712962962963e-05, | |
| "loss": 0.8097, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.4580152671755725, | |
| "grad_norm": 4.816994667053223, | |
| "learning_rate": 3.0285493827160495e-05, | |
| "loss": 0.7754, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.46495489243580845, | |
| "grad_norm": 2.7212016582489014, | |
| "learning_rate": 2.9899691358024696e-05, | |
| "loss": 0.7264, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.4718945176960444, | |
| "grad_norm": 2.6413140296936035, | |
| "learning_rate": 2.951388888888889e-05, | |
| "loss": 0.7636, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.47883414295628035, | |
| "grad_norm": 3.3349661827087402, | |
| "learning_rate": 2.9128086419753087e-05, | |
| "loss": 0.7971, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.4857737682165163, | |
| "grad_norm": 3.037588119506836, | |
| "learning_rate": 2.8780864197530867e-05, | |
| "loss": 0.7778, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.4857737682165163, | |
| "eval_loss": 0.7744709849357605, | |
| "eval_runtime": 50.6673, | |
| "eval_samples_per_second": 28.44, | |
| "eval_steps_per_second": 7.125, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.49271339347675225, | |
| "grad_norm": 3.02775239944458, | |
| "learning_rate": 2.839506172839506e-05, | |
| "loss": 0.7377, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.4996530187369882, | |
| "grad_norm": 3.7178709506988525, | |
| "learning_rate": 2.8009259259259263e-05, | |
| "loss": 0.7315, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.5065926439972241, | |
| "grad_norm": 2.3943405151367188, | |
| "learning_rate": 2.762345679012346e-05, | |
| "loss": 0.6843, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.5135322692574601, | |
| "grad_norm": 3.1974751949310303, | |
| "learning_rate": 2.7237654320987654e-05, | |
| "loss": 0.7292, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.520471894517696, | |
| "grad_norm": 2.935885429382324, | |
| "learning_rate": 2.6851851851851855e-05, | |
| "loss": 0.8194, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.527411519777932, | |
| "grad_norm": 6.58482027053833, | |
| "learning_rate": 2.6466049382716053e-05, | |
| "loss": 0.6681, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.5343511450381679, | |
| "grad_norm": 2.9752790927886963, | |
| "learning_rate": 2.6080246913580247e-05, | |
| "loss": 0.8489, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.5412907702984039, | |
| "grad_norm": 3.415062427520752, | |
| "learning_rate": 2.5694444444444445e-05, | |
| "loss": 0.7535, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.5482303955586398, | |
| "grad_norm": 2.9602854251861572, | |
| "learning_rate": 2.5308641975308646e-05, | |
| "loss": 0.7386, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.5551700208188758, | |
| "grad_norm": 2.3397035598754883, | |
| "learning_rate": 2.492283950617284e-05, | |
| "loss": 0.6889, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.5551700208188758, | |
| "eval_loss": 0.7451881766319275, | |
| "eval_runtime": 50.5382, | |
| "eval_samples_per_second": 28.513, | |
| "eval_steps_per_second": 7.143, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.5621096460791117, | |
| "grad_norm": 1.889366626739502, | |
| "learning_rate": 2.4537037037037038e-05, | |
| "loss": 0.7376, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.5690492713393477, | |
| "grad_norm": 3.3545074462890625, | |
| "learning_rate": 2.4151234567901235e-05, | |
| "loss": 0.7324, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.5759888965995836, | |
| "grad_norm": 3.243748426437378, | |
| "learning_rate": 2.3765432098765433e-05, | |
| "loss": 0.7056, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.5829285218598196, | |
| "grad_norm": 1.6572705507278442, | |
| "learning_rate": 2.337962962962963e-05, | |
| "loss": 0.7623, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.5898681471200555, | |
| "grad_norm": 3.6564319133758545, | |
| "learning_rate": 2.2993827160493828e-05, | |
| "loss": 0.8021, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.5968077723802915, | |
| "grad_norm": 2.7752833366394043, | |
| "learning_rate": 2.2608024691358026e-05, | |
| "loss": 0.6935, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.6037473976405274, | |
| "grad_norm": 4.253413677215576, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.762, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.6106870229007634, | |
| "grad_norm": 2.7186923027038574, | |
| "learning_rate": 2.183641975308642e-05, | |
| "loss": 0.7403, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.6176266481609993, | |
| "grad_norm": 2.6007790565490723, | |
| "learning_rate": 2.145061728395062e-05, | |
| "loss": 0.7119, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.6245662734212353, | |
| "grad_norm": 3.293909788131714, | |
| "learning_rate": 2.1064814814814816e-05, | |
| "loss": 0.6942, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.6245662734212353, | |
| "eval_loss": 0.7264400720596313, | |
| "eval_runtime": 50.4954, | |
| "eval_samples_per_second": 28.537, | |
| "eval_steps_per_second": 7.149, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.6315058986814712, | |
| "grad_norm": 3.3802788257598877, | |
| "learning_rate": 2.0679012345679014e-05, | |
| "loss": 0.7141, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.6384455239417072, | |
| "grad_norm": 2.965416669845581, | |
| "learning_rate": 2.029320987654321e-05, | |
| "loss": 0.7806, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.6453851492019431, | |
| "grad_norm": 3.0134222507476807, | |
| "learning_rate": 1.990740740740741e-05, | |
| "loss": 0.6889, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.6523247744621791, | |
| "grad_norm": 3.078749895095825, | |
| "learning_rate": 1.9521604938271607e-05, | |
| "loss": 0.753, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.659264399722415, | |
| "grad_norm": 2.0064494609832764, | |
| "learning_rate": 1.91358024691358e-05, | |
| "loss": 0.6586, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.666204024982651, | |
| "grad_norm": 3.144228458404541, | |
| "learning_rate": 1.8750000000000002e-05, | |
| "loss": 0.7233, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.6731436502428869, | |
| "grad_norm": 3.3552615642547607, | |
| "learning_rate": 1.83641975308642e-05, | |
| "loss": 0.6848, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 0.6800832755031229, | |
| "grad_norm": 2.424555540084839, | |
| "learning_rate": 1.7978395061728397e-05, | |
| "loss": 0.711, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.6870229007633588, | |
| "grad_norm": 3.5036256313323975, | |
| "learning_rate": 1.7592592592592595e-05, | |
| "loss": 0.7794, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 0.6939625260235948, | |
| "grad_norm": 3.1521127223968506, | |
| "learning_rate": 1.720679012345679e-05, | |
| "loss": 0.7769, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6939625260235948, | |
| "eval_loss": 0.705424964427948, | |
| "eval_runtime": 50.3767, | |
| "eval_samples_per_second": 28.604, | |
| "eval_steps_per_second": 7.166, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7009021512838307, | |
| "grad_norm": 3.177053689956665, | |
| "learning_rate": 1.682098765432099e-05, | |
| "loss": 0.7342, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 0.7078417765440667, | |
| "grad_norm": 2.742203712463379, | |
| "learning_rate": 1.6435185185185187e-05, | |
| "loss": 0.6258, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 0.7147814018043026, | |
| "grad_norm": 2.73271107673645, | |
| "learning_rate": 1.604938271604938e-05, | |
| "loss": 0.7548, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 0.7217210270645386, | |
| "grad_norm": 2.823657751083374, | |
| "learning_rate": 1.5663580246913583e-05, | |
| "loss": 0.7083, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 0.7286606523247745, | |
| "grad_norm": 2.787195920944214, | |
| "learning_rate": 1.527777777777778e-05, | |
| "loss": 0.6906, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.7356002775850105, | |
| "grad_norm": 3.4323344230651855, | |
| "learning_rate": 1.4891975308641976e-05, | |
| "loss": 0.7352, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 0.7425399028452464, | |
| "grad_norm": 2.84405779838562, | |
| "learning_rate": 1.4506172839506174e-05, | |
| "loss": 0.7109, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 0.7494795281054824, | |
| "grad_norm": 2.3355183601379395, | |
| "learning_rate": 1.412037037037037e-05, | |
| "loss": 0.6481, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 0.7564191533657183, | |
| "grad_norm": 3.446125030517578, | |
| "learning_rate": 1.3734567901234569e-05, | |
| "loss": 0.7126, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 0.7633587786259542, | |
| "grad_norm": 3.240204334259033, | |
| "learning_rate": 1.3348765432098767e-05, | |
| "loss": 0.6816, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.7633587786259542, | |
| "eval_loss": 0.6829991936683655, | |
| "eval_runtime": 50.4945, | |
| "eval_samples_per_second": 28.538, | |
| "eval_steps_per_second": 7.149, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.7702984038861902, | |
| "grad_norm": 3.2107372283935547, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 0.6874, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 0.7772380291464261, | |
| "grad_norm": 3.1716368198394775, | |
| "learning_rate": 1.2577160493827162e-05, | |
| "loss": 0.7069, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 0.7841776544066621, | |
| "grad_norm": 2.7112536430358887, | |
| "learning_rate": 1.219135802469136e-05, | |
| "loss": 0.6721, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 0.7911172796668979, | |
| "grad_norm": 3.2708733081817627, | |
| "learning_rate": 1.1805555555555555e-05, | |
| "loss": 0.6656, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 0.7980569049271339, | |
| "grad_norm": 3.133509874343872, | |
| "learning_rate": 1.1419753086419753e-05, | |
| "loss": 0.6972, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.8049965301873698, | |
| "grad_norm": 2.5478899478912354, | |
| "learning_rate": 1.1033950617283952e-05, | |
| "loss": 0.6708, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 0.8119361554476058, | |
| "grad_norm": 3.2422120571136475, | |
| "learning_rate": 1.0648148148148148e-05, | |
| "loss": 0.6752, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 0.8188757807078417, | |
| "grad_norm": 3.1517410278320312, | |
| "learning_rate": 1.0262345679012346e-05, | |
| "loss": 0.6282, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 0.8258154059680777, | |
| "grad_norm": 2.2257463932037354, | |
| "learning_rate": 9.876543209876543e-06, | |
| "loss": 0.6494, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 0.8327550312283136, | |
| "grad_norm": 3.7597286701202393, | |
| "learning_rate": 9.490740740740741e-06, | |
| "loss": 0.7014, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.8327550312283136, | |
| "eval_loss": 0.6637962460517883, | |
| "eval_runtime": 50.4735, | |
| "eval_samples_per_second": 28.55, | |
| "eval_steps_per_second": 7.152, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.8396946564885496, | |
| "grad_norm": 2.7648439407348633, | |
| "learning_rate": 9.104938271604939e-06, | |
| "loss": 0.6734, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 0.8466342817487855, | |
| "grad_norm": 2.5509233474731445, | |
| "learning_rate": 8.719135802469136e-06, | |
| "loss": 0.5945, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 0.8535739070090215, | |
| "grad_norm": 2.9669878482818604, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.6981, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 0.8605135322692574, | |
| "grad_norm": 2.990398406982422, | |
| "learning_rate": 7.947530864197531e-06, | |
| "loss": 0.6788, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 0.8674531575294934, | |
| "grad_norm": 3.9718501567840576, | |
| "learning_rate": 7.561728395061729e-06, | |
| "loss": 0.6359, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.8743927827897293, | |
| "grad_norm": 2.4753811359405518, | |
| "learning_rate": 7.1759259259259266e-06, | |
| "loss": 0.6175, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 0.8813324080499653, | |
| "grad_norm": 2.3380069732666016, | |
| "learning_rate": 6.790123456790123e-06, | |
| "loss": 0.6573, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 0.8882720333102012, | |
| "grad_norm": 2.903273344039917, | |
| "learning_rate": 6.404320987654322e-06, | |
| "loss": 0.6507, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 0.8952116585704372, | |
| "grad_norm": 2.6252002716064453, | |
| "learning_rate": 6.0185185185185185e-06, | |
| "loss": 0.7027, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 0.9021512838306731, | |
| "grad_norm": 2.3435378074645996, | |
| "learning_rate": 5.632716049382716e-06, | |
| "loss": 0.5784, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.9021512838306731, | |
| "eval_loss": 0.6478676199913025, | |
| "eval_runtime": 50.172, | |
| "eval_samples_per_second": 28.721, | |
| "eval_steps_per_second": 7.195, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 2.5476410388946533, | |
| "learning_rate": 5.246913580246914e-06, | |
| "loss": 0.6038, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 0.916030534351145, | |
| "grad_norm": 3.113853931427002, | |
| "learning_rate": 4.861111111111111e-06, | |
| "loss": 0.6613, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 0.922970159611381, | |
| "grad_norm": 1.9195737838745117, | |
| "learning_rate": 4.475308641975309e-06, | |
| "loss": 0.6073, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 0.9299097848716169, | |
| "grad_norm": 3.406034231185913, | |
| "learning_rate": 4.0895061728395066e-06, | |
| "loss": 0.6398, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 0.9368494101318529, | |
| "grad_norm": 2.8904383182525635, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.6691, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.9437890353920888, | |
| "grad_norm": 2.453171968460083, | |
| "learning_rate": 3.3179012345679013e-06, | |
| "loss": 0.6236, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 0.9507286606523248, | |
| "grad_norm": 2.6889538764953613, | |
| "learning_rate": 2.932098765432099e-06, | |
| "loss": 0.6576, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 0.9576682859125607, | |
| "grad_norm": 2.621765613555908, | |
| "learning_rate": 2.546296296296296e-06, | |
| "loss": 0.6016, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 0.9646079111727967, | |
| "grad_norm": 2.4071123600006104, | |
| "learning_rate": 2.1604938271604937e-06, | |
| "loss": 0.6861, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 0.9715475364330326, | |
| "grad_norm": 2.3079018592834473, | |
| "learning_rate": 1.7746913580246913e-06, | |
| "loss": 0.5638, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.9715475364330326, | |
| "eval_loss": 0.6364541053771973, | |
| "eval_runtime": 50.2347, | |
| "eval_samples_per_second": 28.685, | |
| "eval_steps_per_second": 7.186, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.9784871616932685, | |
| "grad_norm": 3.5345683097839355, | |
| "learning_rate": 1.388888888888889e-06, | |
| "loss": 0.6383, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 0.9854267869535045, | |
| "grad_norm": 2.898383140563965, | |
| "learning_rate": 1.0030864197530864e-06, | |
| "loss": 0.6235, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 0.9923664122137404, | |
| "grad_norm": 3.0434718132019043, | |
| "learning_rate": 6.17283950617284e-07, | |
| "loss": 0.669, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 0.9993060374739764, | |
| "grad_norm": 2.3611180782318115, | |
| "learning_rate": 2.3148148148148148e-07, | |
| "loss": 0.5936, | |
| "step": 1440 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1441, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.9964632564865434e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |