| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.9982668977469671, |
| "eval_steps": 500, |
| "global_step": 288, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.01733102253032929, |
| "grad_norm": 19.921022415161133, |
| "learning_rate": 7.904761904761904e-06, |
| "loss": 1.9953, |
| "mean_token_accuracy": 0.5539645373821258, |
| "num_tokens": 880436.0, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.03466204506065858, |
| "grad_norm": 3.1659767627716064, |
| "learning_rate": 1.7785714285714282e-05, |
| "loss": 0.4972, |
| "mean_token_accuracy": 0.8168690085411072, |
| "num_tokens": 1772847.0, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.05199306759098787, |
| "grad_norm": 0.5697739720344543, |
| "learning_rate": 2.7666666666666667e-05, |
| "loss": 0.1821, |
| "mean_token_accuracy": 0.9385778725147247, |
| "num_tokens": 2629531.0, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.06932409012131716, |
| "grad_norm": 0.8860486149787903, |
| "learning_rate": 3.7547619047619045e-05, |
| "loss": 0.1503, |
| "mean_token_accuracy": 0.9442744374275207, |
| "num_tokens": 3524533.0, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.08665511265164645, |
| "grad_norm": 0.44097742438316345, |
| "learning_rate": 4.7428571428571427e-05, |
| "loss": 0.134, |
| "mean_token_accuracy": 0.949429202079773, |
| "num_tokens": 4410812.0, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.10398613518197573, |
| "grad_norm": 0.46751534938812256, |
| "learning_rate": 5.730952380952381e-05, |
| "loss": 0.1097, |
| "mean_token_accuracy": 0.9614084720611572, |
| "num_tokens": 5270909.0, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.12131715771230503, |
| "grad_norm": 0.7552158832550049, |
| "learning_rate": 6.719047619047618e-05, |
| "loss": 0.1009, |
| "mean_token_accuracy": 0.9661354303359986, |
| "num_tokens": 6155537.0, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.1386481802426343, |
| "grad_norm": 0.3426048457622528, |
| "learning_rate": 6.913492996222407e-05, |
| "loss": 0.0927, |
| "mean_token_accuracy": 0.9692380607128144, |
| "num_tokens": 7022278.0, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.1559792027729636, |
| "grad_norm": 0.4504334032535553, |
| "learning_rate": 6.900613269644081e-05, |
| "loss": 0.082, |
| "mean_token_accuracy": 0.9722775280475616, |
| "num_tokens": 7888445.0, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.1733102253032929, |
| "grad_norm": 0.3792800009250641, |
| "learning_rate": 6.877878337145818e-05, |
| "loss": 0.0644, |
| "mean_token_accuracy": 0.9791450917720794, |
| "num_tokens": 8739903.0, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.19064124783362218, |
| "grad_norm": 0.4420663118362427, |
| "learning_rate": 6.845375120299803e-05, |
| "loss": 0.0629, |
| "mean_token_accuracy": 0.9785033941268921, |
| "num_tokens": 9619078.0, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.20797227036395147, |
| "grad_norm": 0.48077255487442017, |
| "learning_rate": 6.803227887375121e-05, |
| "loss": 0.0619, |
| "mean_token_accuracy": 0.9787447690963745, |
| "num_tokens": 10479136.0, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.22530329289428075, |
| "grad_norm": 0.2977643609046936, |
| "learning_rate": 6.751597778227777e-05, |
| "loss": 0.0582, |
| "mean_token_accuracy": 0.9807714521884918, |
| "num_tokens": 11356129.0, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.24263431542461006, |
| "grad_norm": 0.21869313716888428, |
| "learning_rate": 6.690682188221032e-05, |
| "loss": 0.0525, |
| "mean_token_accuracy": 0.9822983920574189, |
| "num_tokens": 12210168.0, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.25996533795493937, |
| "grad_norm": 0.24062839150428772, |
| "learning_rate": 6.620714013531474e-05, |
| "loss": 0.0507, |
| "mean_token_accuracy": 0.9837425649166107, |
| "num_tokens": 13074272.0, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.2772963604852686, |
| "grad_norm": 0.2703610062599182, |
| "learning_rate": 6.541960760726255e-05, |
| "loss": 0.0532, |
| "mean_token_accuracy": 0.9837726652622223, |
| "num_tokens": 13951028.0, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.29462738301559793, |
| "grad_norm": 0.348715603351593, |
| "learning_rate": 6.454723524015746e-05, |
| "loss": 0.0452, |
| "mean_token_accuracy": 0.9850515723228455, |
| "num_tokens": 14853383.0, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.3119584055459272, |
| "grad_norm": 0.23803335428237915, |
| "learning_rate": 6.359335834091915e-05, |
| "loss": 0.0472, |
| "mean_token_accuracy": 0.9840892851352692, |
| "num_tokens": 15724507.0, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.3292894280762565, |
| "grad_norm": 0.16920039057731628, |
| "learning_rate": 6.256162382953555e-05, |
| "loss": 0.0451, |
| "mean_token_accuracy": 0.9848921954631805, |
| "num_tokens": 16566607.0, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.3466204506065858, |
| "grad_norm": 0.32039910554885864, |
| "learning_rate": 6.145597629593734e-05, |
| "loss": 0.0511, |
| "mean_token_accuracy": 0.9832945823669433, |
| "num_tokens": 17476269.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.36395147313691506, |
| "grad_norm": 0.2946508824825287, |
| "learning_rate": 6.028064291880252e-05, |
| "loss": 0.0462, |
| "mean_token_accuracy": 0.9852570235729218, |
| "num_tokens": 18336061.0, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.38128249566724437, |
| "grad_norm": 0.18868154287338257, |
| "learning_rate": 5.904011730395065e-05, |
| "loss": 0.0441, |
| "mean_token_accuracy": 0.9859723746776581, |
| "num_tokens": 19197400.0, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.3986135181975737, |
| "grad_norm": 0.3599472939968109, |
| "learning_rate": 5.7739142304116696e-05, |
| "loss": 0.0402, |
| "mean_token_accuracy": 0.9861063182353973, |
| "num_tokens": 20052933.0, |
| "step": 115 |
| }, |
| { |
| "epoch": 0.41594454072790293, |
| "grad_norm": 0.2798216640949249, |
| "learning_rate": 5.63826918857889e-05, |
| "loss": 0.0389, |
| "mean_token_accuracy": 0.9861369669437409, |
| "num_tokens": 20928177.0, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.43327556325823224, |
| "grad_norm": 0.38542649149894714, |
| "learning_rate": 5.4975952112438574e-05, |
| "loss": 0.0341, |
| "mean_token_accuracy": 0.9876367330551148, |
| "num_tokens": 21793162.0, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.4506065857885615, |
| "grad_norm": 0.2921411991119385, |
| "learning_rate": 5.35243013168478e-05, |
| "loss": 0.0359, |
| "mean_token_accuracy": 0.9860971868038177, |
| "num_tokens": 22681479.0, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.4679376083188908, |
| "grad_norm": 0.19987064599990845, |
| "learning_rate": 5.203328953834159e-05, |
| "loss": 0.033, |
| "mean_token_accuracy": 0.9898090362548828, |
| "num_tokens": 23536222.0, |
| "step": 135 |
| }, |
| { |
| "epoch": 0.4852686308492201, |
| "grad_norm": 0.3248032033443451, |
| "learning_rate": 5.050861730354096e-05, |
| "loss": 0.0305, |
| "mean_token_accuracy": 0.9885973572731018, |
| "num_tokens": 24414936.0, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.5025996533795494, |
| "grad_norm": 0.21195006370544434, |
| "learning_rate": 4.895611383176404e-05, |
| "loss": 0.0346, |
| "mean_token_accuracy": 0.9883223176002502, |
| "num_tokens": 25296910.0, |
| "step": 145 |
| }, |
| { |
| "epoch": 0.5199306759098787, |
| "grad_norm": 0.23912712931632996, |
| "learning_rate": 4.7381714748401414e-05, |
| "loss": 0.0289, |
| "mean_token_accuracy": 0.9890963315963746, |
| "num_tokens": 26140083.0, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.537261698440208, |
| "grad_norm": 0.22459208965301514, |
| "learning_rate": 4.579143939147321e-05, |
| "loss": 0.0281, |
| "mean_token_accuracy": 0.9906902074813843, |
| "num_tokens": 27000947.0, |
| "step": 155 |
| }, |
| { |
| "epoch": 0.5545927209705372, |
| "grad_norm": 0.28445667028427124, |
| "learning_rate": 4.4191367798131114e-05, |
| "loss": 0.0291, |
| "mean_token_accuracy": 0.9899352192878723, |
| "num_tokens": 27877835.0, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.5719237435008665, |
| "grad_norm": 0.24794110655784607, |
| "learning_rate": 4.2587617459091795e-05, |
| "loss": 0.027, |
| "mean_token_accuracy": 0.9906977355480194, |
| "num_tokens": 28750062.0, |
| "step": 165 |
| }, |
| { |
| "epoch": 0.5892547660311959, |
| "grad_norm": 0.2798149287700653, |
| "learning_rate": 4.0986319929875456e-05, |
| "loss": 0.0214, |
| "mean_token_accuracy": 0.9938373386859893, |
| "num_tokens": 29584073.0, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.6065857885615251, |
| "grad_norm": 0.2256922572851181, |
| "learning_rate": 3.939359738827107e-05, |
| "loss": 0.0274, |
| "mean_token_accuracy": 0.9910281479358674, |
| "num_tokens": 30493103.0, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.6239168110918544, |
| "grad_norm": 0.18433739244937897, |
| "learning_rate": 3.781553922765516e-05, |
| "loss": 0.0266, |
| "mean_token_accuracy": 0.9907304227352143, |
| "num_tokens": 31367886.0, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.6412478336221837, |
| "grad_norm": 0.17792819440364838, |
| "learning_rate": 3.6258178775653865e-05, |
| "loss": 0.0229, |
| "mean_token_accuracy": 0.9927768170833587, |
| "num_tokens": 32241872.0, |
| "step": 185 |
| }, |
| { |
| "epoch": 0.658578856152513, |
| "grad_norm": 0.2896949052810669, |
| "learning_rate": 3.4727470227158956e-05, |
| "loss": 0.0256, |
| "mean_token_accuracy": 0.9920165657997131, |
| "num_tokens": 33090146.0, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.6759098786828422, |
| "grad_norm": 0.15475758910179138, |
| "learning_rate": 3.322926587988904e-05, |
| "loss": 0.0198, |
| "mean_token_accuracy": 0.9928617775440216, |
| "num_tokens": 33946268.0, |
| "step": 195 |
| }, |
| { |
| "epoch": 0.6932409012131716, |
| "grad_norm": 0.17668993771076202, |
| "learning_rate": 3.1769293759530075e-05, |
| "loss": 0.0196, |
| "mean_token_accuracy": 0.9939430177211761, |
| "num_tokens": 34838335.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.7105719237435009, |
| "grad_norm": 0.1672198474407196, |
| "learning_rate": 3.035313572000036e-05, |
| "loss": 0.019, |
| "mean_token_accuracy": 0.9942795395851135, |
| "num_tokens": 35696643.0, |
| "step": 205 |
| }, |
| { |
| "epoch": 0.7279029462738301, |
| "grad_norm": 0.1975901871919632, |
| "learning_rate": 2.898620610256823e-05, |
| "loss": 0.0232, |
| "mean_token_accuracy": 0.9921872437000274, |
| "num_tokens": 36594681.0, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.7452339688041595, |
| "grad_norm": 0.191142275929451, |
| "learning_rate": 2.767373103541428e-05, |
| "loss": 0.0212, |
| "mean_token_accuracy": 0.9928397178649903, |
| "num_tokens": 37463957.0, |
| "step": 215 |
| }, |
| { |
| "epoch": 0.7625649913344887, |
| "grad_norm": 0.20682832598686218, |
| "learning_rate": 2.6420728452781278e-05, |
| "loss": 0.0199, |
| "mean_token_accuracy": 0.9933333337306977, |
| "num_tokens": 38357527.0, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.779896013864818, |
| "grad_norm": 0.2716505527496338, |
| "learning_rate": 2.523198891010382e-05, |
| "loss": 0.0222, |
| "mean_token_accuracy": 0.9922467291355133, |
| "num_tokens": 39241233.0, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.7972270363951474, |
| "grad_norm": 0.19001668691635132, |
| "learning_rate": 2.4112057268466295e-05, |
| "loss": 0.0209, |
| "mean_token_accuracy": 0.9923349857330322, |
| "num_tokens": 40109881.0, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.8145580589254766, |
| "grad_norm": 0.29096758365631104, |
| "learning_rate": 2.3065215318414493e-05, |
| "loss": 0.0256, |
| "mean_token_accuracy": 0.9918012678623199, |
| "num_tokens": 40989898.0, |
| "step": 235 |
| }, |
| { |
| "epoch": 0.8318890814558059, |
| "grad_norm": 0.18268224596977234, |
| "learning_rate": 2.2095465409554474e-05, |
| "loss": 0.0184, |
| "mean_token_accuracy": 0.9939315497875214, |
| "num_tokens": 41856051.0, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.8492201039861352, |
| "grad_norm": 0.25573816895484924, |
| "learning_rate": 2.1206515148526813e-05, |
| "loss": 0.0196, |
| "mean_token_accuracy": 0.9933596789836884, |
| "num_tokens": 42735924.0, |
| "step": 245 |
| }, |
| { |
| "epoch": 0.8665511265164645, |
| "grad_norm": 0.1488819718360901, |
| "learning_rate": 2.040176322386022e-05, |
| "loss": 0.016, |
| "mean_token_accuracy": 0.9949212014675141, |
| "num_tokens": 43615669.0, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.8838821490467937, |
| "grad_norm": 0.1831093579530716, |
| "learning_rate": 1.9684286411899473e-05, |
| "loss": 0.0166, |
| "mean_token_accuracy": 0.9946038603782654, |
| "num_tokens": 44503843.0, |
| "step": 255 |
| }, |
| { |
| "epoch": 0.901213171577123, |
| "grad_norm": 0.14144335687160492, |
| "learning_rate": 1.9056827813487584e-05, |
| "loss": 0.0147, |
| "mean_token_accuracy": 0.9956230521202087, |
| "num_tokens": 45376996.0, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.9185441941074524, |
| "grad_norm": 0.15868957340717316, |
| "learning_rate": 1.8521786366376354e-05, |
| "loss": 0.0154, |
| "mean_token_accuracy": 0.9951422095298768, |
| "num_tokens": 46247841.0, |
| "step": 265 |
| }, |
| { |
| "epoch": 0.9358752166377816, |
| "grad_norm": 0.19115042686462402, |
| "learning_rate": 1.8081207673462145e-05, |
| "loss": 0.0166, |
| "mean_token_accuracy": 0.9950133621692657, |
| "num_tokens": 47119713.0, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.9532062391681109, |
| "grad_norm": 0.25370219349861145, |
| "learning_rate": 1.7736776181912865e-05, |
| "loss": 0.0178, |
| "mean_token_accuracy": 0.9936381220817566, |
| "num_tokens": 48016981.0, |
| "step": 275 |
| }, |
| { |
| "epoch": 0.9705372616984402, |
| "grad_norm": 0.15300053358078003, |
| "learning_rate": 1.7489808743087434e-05, |
| "loss": 0.0167, |
| "mean_token_accuracy": 0.9948623299598693, |
| "num_tokens": 48887308.0, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.9878682842287695, |
| "grad_norm": 0.16295602917671204, |
| "learning_rate": 1.7341249577869848e-05, |
| "loss": 0.0179, |
| "mean_token_accuracy": 0.9947715997695923, |
| "num_tokens": 49749889.0, |
| "step": 285 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 289, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "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": 2.6996040282155254e+18, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
