| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.600925862789154, | |
| "learning_rate": 4e-05, | |
| "loss": 0.4307, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.6059685349464417, | |
| "learning_rate": 8e-05, | |
| "loss": 0.4353, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.5777328014373779, | |
| "learning_rate": 0.00012, | |
| "loss": 0.421, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.548288106918335, | |
| "learning_rate": 0.00016, | |
| "loss": 0.4029, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.4978249967098236, | |
| "learning_rate": 0.0002, | |
| "loss": 0.4102, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.5470420718193054, | |
| "learning_rate": 0.00019994532573409262, | |
| "loss": 0.4123, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.5333985686302185, | |
| "learning_rate": 0.00019978136272187747, | |
| "loss": 0.3966, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.4586232304573059, | |
| "learning_rate": 0.00019950829025450114, | |
| "loss": 0.377, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.49000784754753113, | |
| "learning_rate": 0.00019912640693269752, | |
| "loss": 0.3785, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.45979875326156616, | |
| "learning_rate": 0.00019863613034027224, | |
| "loss": 0.3957, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.4552418887615204, | |
| "learning_rate": 0.00019803799658748094, | |
| "loss": 0.3921, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.4663061201572418, | |
| "learning_rate": 0.0001973326597248006, | |
| "loss": 0.3896, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.4285649359226227, | |
| "learning_rate": 0.00019652089102773488, | |
| "loss": 0.349, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.45254892110824585, | |
| "learning_rate": 0.00019560357815343577, | |
| "loss": 0.3685, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.4699627459049225, | |
| "learning_rate": 0.00019458172417006347, | |
| "loss": 0.3765, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.4629937708377838, | |
| "learning_rate": 0.0001934564464599461, | |
| "loss": 0.3567, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.44340917468070984, | |
| "learning_rate": 0.00019222897549773848, | |
| "loss": 0.3742, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.48120179772377014, | |
| "learning_rate": 0.00019090065350491626, | |
| "loss": 0.3507, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.4626652002334595, | |
| "learning_rate": 0.00018947293298207635, | |
| "loss": 0.3637, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.4551631808280945, | |
| "learning_rate": 0.0001879473751206489, | |
| "loss": 0.3588, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.47978681325912476, | |
| "learning_rate": 0.00018632564809575742, | |
| "loss": 0.3631, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.44357830286026, | |
| "learning_rate": 0.00018460952524209355, | |
| "loss": 0.3634, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.4979308247566223, | |
| "learning_rate": 0.00018280088311480201, | |
| "loss": 0.3497, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.4785916805267334, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.3506, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.45637139678001404, | |
| "learning_rate": 0.00017891405093963938, | |
| "loss": 0.3621, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.4222208261489868, | |
| "learning_rate": 0.00017684011108568592, | |
| "loss": 0.333, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.4473680555820465, | |
| "learning_rate": 0.0001746821476984154, | |
| "loss": 0.3384, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.45950424671173096, | |
| "learning_rate": 0.00017244252047910892, | |
| "loss": 0.3396, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.4503507614135742, | |
| "learning_rate": 0.00017012367842724887, | |
| "loss": 0.3907, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.4551246166229248, | |
| "learning_rate": 0.00016772815716257412, | |
| "loss": 0.3413, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.42266467213630676, | |
| "learning_rate": 0.00016525857615241687, | |
| "loss": 0.3402, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.473186194896698, | |
| "learning_rate": 0.0001627176358473537, | |
| "loss": 0.3515, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.42668187618255615, | |
| "learning_rate": 0.00016010811472830252, | |
| "loss": 0.3148, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.424800306558609, | |
| "learning_rate": 0.00015743286626829437, | |
| "loss": 0.3224, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.43106356263160706, | |
| "learning_rate": 0.00015469481581224272, | |
| "loss": 0.3143, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.4076085090637207, | |
| "learning_rate": 0.00015189695737812152, | |
| "loss": 0.3329, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.4057788550853729, | |
| "learning_rate": 0.00014904235038305083, | |
| "loss": 0.3389, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.4162575900554657, | |
| "learning_rate": 0.0001461341162978688, | |
| "loss": 0.3268, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.42211902141571045, | |
| "learning_rate": 0.00014317543523384928, | |
| "loss": 0.2971, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.42816001176834106, | |
| "learning_rate": 0.00014016954246529696, | |
| "loss": 0.3303, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.4197749197483063, | |
| "learning_rate": 0.00013711972489182208, | |
| "loss": 0.3443, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.41594424843788147, | |
| "learning_rate": 0.00013402931744416433, | |
| "loss": 0.3417, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.41279336810112, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.3252, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.42070260643959045, | |
| "learning_rate": 0.00012774029087618446, | |
| "loss": 0.3107, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.4160784184932709, | |
| "learning_rate": 0.00012454854871407994, | |
| "loss": 0.3253, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.4144185483455658, | |
| "learning_rate": 0.0001213299630743747, | |
| "loss": 0.3251, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.3940325081348419, | |
| "learning_rate": 0.000118088053433211, | |
| "loss": 0.3234, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.40852218866348267, | |
| "learning_rate": 0.0001148263647711842, | |
| "loss": 0.3405, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 0.4145854711532593, | |
| "learning_rate": 0.00011154846369695863, | |
| "loss": 0.3237, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.40923136472702026, | |
| "learning_rate": 0.00010825793454723325, | |
| "loss": 0.3401, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 0.33250558376312256, | |
| "learning_rate": 0.00010495837546732224, | |
| "loss": 0.2383, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 0.3187699019908905, | |
| "learning_rate": 0.00010165339447663587, | |
| "loss": 0.2129, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 0.3277425467967987, | |
| "learning_rate": 9.834660552336415e-05, | |
| "loss": 0.207, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 0.33040741086006165, | |
| "learning_rate": 9.504162453267777e-05, | |
| "loss": 0.1989, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 0.33165445923805237, | |
| "learning_rate": 9.174206545276677e-05, | |
| "loss": 0.224, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 0.3344186842441559, | |
| "learning_rate": 8.845153630304139e-05, | |
| "loss": 0.2011, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.1400000000000001, | |
| "grad_norm": 0.3419267535209656, | |
| "learning_rate": 8.517363522881579e-05, | |
| "loss": 0.2195, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 0.3256666362285614, | |
| "learning_rate": 8.191194656678904e-05, | |
| "loss": 0.1901, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 0.3259730935096741, | |
| "learning_rate": 7.867003692562534e-05, | |
| "loss": 0.2043, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.3260168433189392, | |
| "learning_rate": 7.54514512859201e-05, | |
| "loss": 0.2002, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "grad_norm": 0.34458276629447937, | |
| "learning_rate": 7.225970912381556e-05, | |
| "loss": 0.2031, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 0.3367977440357208, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.2018, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 0.36229151487350464, | |
| "learning_rate": 6.59706825558357e-05, | |
| "loss": 0.1987, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 0.34255215525627136, | |
| "learning_rate": 6.28802751081779e-05, | |
| "loss": 0.1906, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 0.3520965576171875, | |
| "learning_rate": 5.983045753470308e-05, | |
| "loss": 0.1916, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 0.3631666600704193, | |
| "learning_rate": 5.6824564766150726e-05, | |
| "loss": 0.2055, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "grad_norm": 0.3751111626625061, | |
| "learning_rate": 5.386588370213124e-05, | |
| "loss": 0.1894, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 1.3599999999999999, | |
| "grad_norm": 0.3539924919605255, | |
| "learning_rate": 5.095764961694922e-05, | |
| "loss": 0.1949, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "grad_norm": 0.35575294494628906, | |
| "learning_rate": 4.810304262187852e-05, | |
| "loss": 0.1947, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.3652094602584839, | |
| "learning_rate": 4.530518418775733e-05, | |
| "loss": 0.1937, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 0.3642243444919586, | |
| "learning_rate": 4.256713373170564e-05, | |
| "loss": 0.2066, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.37013036012649536, | |
| "learning_rate": 3.9891885271697496e-05, | |
| "loss": 0.2062, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "grad_norm": 0.35074442625045776, | |
| "learning_rate": 3.7282364152646297e-05, | |
| "loss": 0.2015, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 0.35445520281791687, | |
| "learning_rate": 3.4741423847583134e-05, | |
| "loss": 0.1934, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.3767736852169037, | |
| "learning_rate": 3.227184283742591e-05, | |
| "loss": 0.1955, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 0.3581729829311371, | |
| "learning_rate": 2.9876321572751144e-05, | |
| "loss": 0.1882, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 0.362015038728714, | |
| "learning_rate": 2.7557479520891104e-05, | |
| "loss": 0.2056, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 0.35143327713012695, | |
| "learning_rate": 2.5317852301584643e-05, | |
| "loss": 0.1957, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 0.3580865263938904, | |
| "learning_rate": 2.315988891431412e-05, | |
| "loss": 0.197, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.3541586399078369, | |
| "learning_rate": 2.1085949060360654e-05, | |
| "loss": 0.2022, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "grad_norm": 0.36059048771858215, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.1855, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "grad_norm": 0.3643834888935089, | |
| "learning_rate": 1.7199116885197995e-05, | |
| "loss": 0.1812, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 1.6600000000000001, | |
| "grad_norm": 0.3523213863372803, | |
| "learning_rate": 1.5390474757906446e-05, | |
| "loss": 0.198, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 1.6800000000000002, | |
| "grad_norm": 0.3479655981063843, | |
| "learning_rate": 1.3674351904242611e-05, | |
| "loss": 0.1913, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 0.3516225218772888, | |
| "learning_rate": 1.2052624879351104e-05, | |
| "loss": 0.1936, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 0.3603803217411041, | |
| "learning_rate": 1.0527067017923654e-05, | |
| "loss": 0.182, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "grad_norm": 0.3666897416114807, | |
| "learning_rate": 9.09934649508375e-06, | |
| "loss": 0.198, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 0.347070574760437, | |
| "learning_rate": 7.771024502261526e-06, | |
| "loss": 0.1819, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "grad_norm": 0.366550087928772, | |
| "learning_rate": 6.543553540053926e-06, | |
| "loss": 0.1958, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 0.3641619384288788, | |
| "learning_rate": 5.418275829936537e-06, | |
| "loss": 0.193, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.8199999999999998, | |
| "grad_norm": 0.3448559045791626, | |
| "learning_rate": 4.3964218465642355e-06, | |
| "loss": 0.183, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 1.8399999999999999, | |
| "grad_norm": 0.3684370517730713, | |
| "learning_rate": 3.4791089722651436e-06, | |
| "loss": 0.2036, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 1.8599999999999999, | |
| "grad_norm": 0.3479262590408325, | |
| "learning_rate": 2.667340275199426e-06, | |
| "loss": 0.1871, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 0.3530682325363159, | |
| "learning_rate": 1.9620034125190644e-06, | |
| "loss": 0.1975, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 0.3446263074874878, | |
| "learning_rate": 1.3638696597277679e-06, | |
| "loss": 0.1791, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.341688334941864, | |
| "learning_rate": 8.735930673024806e-07, | |
| "loss": 0.1954, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "grad_norm": 0.3435825705528259, | |
| "learning_rate": 4.917097454988584e-07, | |
| "loss": 0.1872, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 0.35581129789352417, | |
| "learning_rate": 2.1863727812254653e-07, | |
| "loss": 0.1927, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 0.362829327583313, | |
| "learning_rate": 5.467426590739511e-08, | |
| "loss": 0.1888, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.35802093148231506, | |
| "learning_rate": 0.0, | |
| "loss": 0.1933, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.504376951971676e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |