| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9932279909706545, | |
| "eval_steps": 50, | |
| "global_step": 663, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.022573363431151242, | |
| "grad_norm": 18.940821260700204, | |
| "learning_rate": 5e-07, | |
| "loss": 1.7274, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.045146726862302484, | |
| "grad_norm": 12.069870592838273, | |
| "learning_rate": 1e-06, | |
| "loss": 1.5904, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06772009029345373, | |
| "grad_norm": 6.898216686215923, | |
| "learning_rate": 9.998553453302385e-07, | |
| "loss": 1.2673, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.09029345372460497, | |
| "grad_norm": 3.9798223515798434, | |
| "learning_rate": 9.99421465020848e-07, | |
| "loss": 1.1381, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.11286681715575621, | |
| "grad_norm": 4.010503230658132, | |
| "learning_rate": 9.9869861012308e-07, | |
| "loss": 1.068, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.13544018058690746, | |
| "grad_norm": 3.8938790644805055, | |
| "learning_rate": 9.976871988942804e-07, | |
| "loss": 1.0093, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1580135440180587, | |
| "grad_norm": 3.670828852459692, | |
| "learning_rate": 9.963878165558785e-07, | |
| "loss": 0.965, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.18058690744920994, | |
| "grad_norm": 3.7471523569522636, | |
| "learning_rate": 9.948012149547666e-07, | |
| "loss": 0.9569, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.20316027088036118, | |
| "grad_norm": 3.6106762024402306, | |
| "learning_rate": 9.929283121282675e-07, | |
| "loss": 0.9574, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.22573363431151242, | |
| "grad_norm": 3.697158068556862, | |
| "learning_rate": 9.9077019177294e-07, | |
| "loss": 0.9348, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.22573363431151242, | |
| "eval_loss": 0.9328124523162842, | |
| "eval_runtime": 110.1622, | |
| "eval_samples_per_second": 57.188, | |
| "eval_steps_per_second": 0.899, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.24830699774266365, | |
| "grad_norm": 4.016927586496979, | |
| "learning_rate": 9.88328102617534e-07, | |
| "loss": 0.9537, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.2708803611738149, | |
| "grad_norm": 3.493701469135057, | |
| "learning_rate": 9.856034577004504e-07, | |
| "loss": 0.911, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.29345372460496616, | |
| "grad_norm": 3.4389902881459276, | |
| "learning_rate": 9.82597833552132e-07, | |
| "loss": 0.9246, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.3160270880361174, | |
| "grad_norm": 3.494290022340395, | |
| "learning_rate": 9.793129692828533e-07, | |
| "loss": 0.8965, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.33860045146726864, | |
| "grad_norm": 3.3273935998770447, | |
| "learning_rate": 9.757507655764384e-07, | |
| "loss": 0.8886, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.3611738148984199, | |
| "grad_norm": 3.507197053253818, | |
| "learning_rate": 9.719132835904906e-07, | |
| "loss": 0.8754, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.3837471783295711, | |
| "grad_norm": 3.650140545940056, | |
| "learning_rate": 9.678027437637677e-07, | |
| "loss": 0.8817, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.40632054176072235, | |
| "grad_norm": 3.37664746916909, | |
| "learning_rate": 9.634215245313939e-07, | |
| "loss": 0.8962, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4288939051918736, | |
| "grad_norm": 3.6123744733886562, | |
| "learning_rate": 9.587721609486543e-07, | |
| "loss": 0.8889, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.45146726862302483, | |
| "grad_norm": 3.7700865037416222, | |
| "learning_rate": 9.538573432241637e-07, | |
| "loss": 0.8794, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.45146726862302483, | |
| "eval_loss": 0.8777003288269043, | |
| "eval_runtime": 109.6583, | |
| "eval_samples_per_second": 57.451, | |
| "eval_steps_per_second": 0.903, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.47404063205417607, | |
| "grad_norm": 3.463064756132834, | |
| "learning_rate": 9.486799151632612e-07, | |
| "loss": 0.889, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.4966139954853273, | |
| "grad_norm": 3.5807866593622797, | |
| "learning_rate": 9.432428725225326e-07, | |
| "loss": 0.8764, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.5191873589164786, | |
| "grad_norm": 3.4420739762541253, | |
| "learning_rate": 9.375493612764085e-07, | |
| "loss": 0.8801, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.5417607223476298, | |
| "grad_norm": 3.398507778758291, | |
| "learning_rate": 9.316026757968454e-07, | |
| "loss": 0.8595, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.5643340857787811, | |
| "grad_norm": 3.540107126680079, | |
| "learning_rate": 9.2540625694714e-07, | |
| "loss": 0.8595, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.5869074492099323, | |
| "grad_norm": 3.5675667181372255, | |
| "learning_rate": 9.189636900909817e-07, | |
| "loss": 0.8596, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.6094808126410836, | |
| "grad_norm": 3.476803495407312, | |
| "learning_rate": 9.122787030178949e-07, | |
| "loss": 0.8682, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.6320541760722348, | |
| "grad_norm": 3.734209530115444, | |
| "learning_rate": 9.053551637862692e-07, | |
| "loss": 0.8663, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.654627539503386, | |
| "grad_norm": 3.780791382608422, | |
| "learning_rate": 8.98197078485229e-07, | |
| "loss": 0.8551, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.6772009029345373, | |
| "grad_norm": 3.3743939042622024, | |
| "learning_rate": 8.908085889166357e-07, | |
| "loss": 0.851, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.6772009029345373, | |
| "eval_loss": 0.8527019023895264, | |
| "eval_runtime": 109.652, | |
| "eval_samples_per_second": 57.454, | |
| "eval_steps_per_second": 0.903, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.6997742663656885, | |
| "grad_norm": 3.7470275771419037, | |
| "learning_rate": 8.831939701985636e-07, | |
| "loss": 0.8529, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.7223476297968398, | |
| "grad_norm": 3.636834432144132, | |
| "learning_rate": 8.75357628291637e-07, | |
| "loss": 0.8348, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.744920993227991, | |
| "grad_norm": 3.452067647076289, | |
| "learning_rate": 8.673040974496584e-07, | |
| "loss": 0.8617, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.7674943566591422, | |
| "grad_norm": 3.526963633154407, | |
| "learning_rate": 8.590380375960053e-07, | |
| "loss": 0.8467, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.7900677200902935, | |
| "grad_norm": 3.673328628232425, | |
| "learning_rate": 8.505642316273111e-07, | |
| "loss": 0.8609, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.8126410835214447, | |
| "grad_norm": 3.5342947955254664, | |
| "learning_rate": 8.418875826459919e-07, | |
| "loss": 0.8252, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.835214446952596, | |
| "grad_norm": 3.7740248810423123, | |
| "learning_rate": 8.330131111232201e-07, | |
| "loss": 0.8514, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.8577878103837472, | |
| "grad_norm": 3.4551265980839605, | |
| "learning_rate": 8.239459519939851e-07, | |
| "loss": 0.8199, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.8803611738148984, | |
| "grad_norm": 3.7207649703953924, | |
| "learning_rate": 8.14691351685925e-07, | |
| "loss": 0.8362, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.9029345372460497, | |
| "grad_norm": 3.507860484341248, | |
| "learning_rate": 8.052546650836453e-07, | |
| "loss": 0.8374, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.9029345372460497, | |
| "eval_loss": 0.8355751633644104, | |
| "eval_runtime": 109.7026, | |
| "eval_samples_per_second": 57.428, | |
| "eval_steps_per_second": 0.902, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.9255079006772009, | |
| "grad_norm": 3.5155465231821035, | |
| "learning_rate": 7.956413524302823e-07, | |
| "loss": 0.8268, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.9480812641083521, | |
| "grad_norm": 3.4314347312225473, | |
| "learning_rate": 7.858569761681047e-07, | |
| "loss": 0.8433, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.9706546275395034, | |
| "grad_norm": 3.513861072295851, | |
| "learning_rate": 7.759071977199806e-07, | |
| "loss": 0.84, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.9932279909706546, | |
| "grad_norm": 3.3821820044041617, | |
| "learning_rate": 7.657977742135725e-07, | |
| "loss": 0.8223, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.0158013544018059, | |
| "grad_norm": 3.7538884588461685, | |
| "learning_rate": 7.555345551501557e-07, | |
| "loss": 0.8016, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.0383747178329572, | |
| "grad_norm": 3.3692234546119555, | |
| "learning_rate": 7.451234790199871e-07, | |
| "loss": 0.7731, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.0609480812641083, | |
| "grad_norm": 3.3350384318615336, | |
| "learning_rate": 7.345705698661852e-07, | |
| "loss": 0.7721, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.0835214446952597, | |
| "grad_norm": 3.378786795745363, | |
| "learning_rate": 7.23881933799104e-07, | |
| "loss": 0.7618, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.1060948081264108, | |
| "grad_norm": 3.7399776397062716, | |
| "learning_rate": 7.130637554632257e-07, | |
| "loss": 0.771, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.1286681715575622, | |
| "grad_norm": 3.5991184360567434, | |
| "learning_rate": 7.021222944586088e-07, | |
| "loss": 0.7654, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.1286681715575622, | |
| "eval_loss": 0.828125, | |
| "eval_runtime": 109.5411, | |
| "eval_samples_per_second": 57.513, | |
| "eval_steps_per_second": 0.904, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.1512415349887133, | |
| "grad_norm": 3.6074769131210913, | |
| "learning_rate": 6.910638817189664e-07, | |
| "loss": 0.7535, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.1738148984198646, | |
| "grad_norm": 3.593924753487917, | |
| "learning_rate": 6.798949158484705e-07, | |
| "loss": 0.7603, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.1963882618510158, | |
| "grad_norm": 3.6842003754580595, | |
| "learning_rate": 6.686218594193993e-07, | |
| "loss": 0.7691, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.2189616252821671, | |
| "grad_norm": 3.5700230474280468, | |
| "learning_rate": 6.572512352327726e-07, | |
| "loss": 0.7697, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.2415349887133182, | |
| "grad_norm": 3.488732480561882, | |
| "learning_rate": 6.457896225441371e-07, | |
| "loss": 0.7534, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.2641083521444696, | |
| "grad_norm": 3.508919159526223, | |
| "learning_rate": 6.342436532566865e-07, | |
| "loss": 0.7697, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.2866817155756207, | |
| "grad_norm": 3.6151399702366893, | |
| "learning_rate": 6.226200080839182e-07, | |
| "loss": 0.7701, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.309255079006772, | |
| "grad_norm": 3.860906370816312, | |
| "learning_rate": 6.109254126840479e-07, | |
| "loss": 0.7731, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.3318284424379232, | |
| "grad_norm": 3.8153392985538126, | |
| "learning_rate": 5.991666337684176e-07, | |
| "loss": 0.758, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.3544018058690745, | |
| "grad_norm": 3.586676854711127, | |
| "learning_rate": 5.873504751861507e-07, | |
| "loss": 0.7551, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.3544018058690745, | |
| "eval_loss": 0.8203195929527283, | |
| "eval_runtime": 109.4763, | |
| "eval_samples_per_second": 57.547, | |
| "eval_steps_per_second": 0.904, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.3769751693002257, | |
| "grad_norm": 3.6095163997213904, | |
| "learning_rate": 5.754837739873178e-07, | |
| "loss": 0.7589, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 1.399548532731377, | |
| "grad_norm": 3.3338115229207927, | |
| "learning_rate": 5.635733964668909e-07, | |
| "loss": 0.7444, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.4221218961625282, | |
| "grad_norm": 3.3851671162659165, | |
| "learning_rate": 5.516262341917778e-07, | |
| "loss": 0.7585, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 1.4446952595936795, | |
| "grad_norm": 3.662781919796072, | |
| "learning_rate": 5.396492000132325e-07, | |
| "loss": 0.7562, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.4672686230248306, | |
| "grad_norm": 3.638999696510764, | |
| "learning_rate": 5.276492240669503e-07, | |
| "loss": 0.7592, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.489841986455982, | |
| "grad_norm": 3.369757400210188, | |
| "learning_rate": 5.156332497631621e-07, | |
| "loss": 0.7585, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.5124153498871333, | |
| "grad_norm": 3.6296512106580776, | |
| "learning_rate": 5.036082297690464e-07, | |
| "loss": 0.751, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 1.5349887133182845, | |
| "grad_norm": 3.536558577874657, | |
| "learning_rate": 4.915811219857882e-07, | |
| "loss": 0.7559, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.5575620767494356, | |
| "grad_norm": 3.8267082837972675, | |
| "learning_rate": 4.795588855226055e-07, | |
| "loss": 0.7553, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 1.580135440180587, | |
| "grad_norm": 3.5316321553516348, | |
| "learning_rate": 4.6754847667008004e-07, | |
| "loss": 0.7532, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.580135440180587, | |
| "eval_loss": 0.814508318901062, | |
| "eval_runtime": 109.6529, | |
| "eval_samples_per_second": 57.454, | |
| "eval_steps_per_second": 0.903, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.6027088036117383, | |
| "grad_norm": 3.559386810490228, | |
| "learning_rate": 4.5555684487511693e-07, | |
| "loss": 0.7446, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 1.6252821670428894, | |
| "grad_norm": 3.930079470266933, | |
| "learning_rate": 4.435909287198646e-07, | |
| "loss": 0.7641, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.6478555304740405, | |
| "grad_norm": 3.5973562045419616, | |
| "learning_rate": 4.316576519069226e-07, | |
| "loss": 0.7355, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 1.670428893905192, | |
| "grad_norm": 3.4517917709956576, | |
| "learning_rate": 4.197639192531573e-07, | |
| "loss": 0.7302, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.6930022573363432, | |
| "grad_norm": 3.496708353360656, | |
| "learning_rate": 4.079166126944453e-07, | |
| "loss": 0.7521, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.7155756207674944, | |
| "grad_norm": 3.622062773262266, | |
| "learning_rate": 3.9612258730365823e-07, | |
| "loss": 0.7496, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.7381489841986455, | |
| "grad_norm": 3.415450213597566, | |
| "learning_rate": 3.843886673241883e-07, | |
| "loss": 0.7311, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 1.7607223476297968, | |
| "grad_norm": 3.541495254850115, | |
| "learning_rate": 3.7272164222131387e-07, | |
| "loss": 0.7497, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.7832957110609482, | |
| "grad_norm": 3.6679244301122638, | |
| "learning_rate": 3.611282627536887e-07, | |
| "loss": 0.7565, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 1.8058690744920993, | |
| "grad_norm": 3.6156298321209754, | |
| "learning_rate": 3.496152370672255e-07, | |
| "loss": 0.7599, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.8058690744920993, | |
| "eval_loss": 0.809232771396637, | |
| "eval_runtime": 109.5418, | |
| "eval_samples_per_second": 57.512, | |
| "eval_steps_per_second": 0.904, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.8284424379232505, | |
| "grad_norm": 3.763934827849069, | |
| "learning_rate": 3.381892268136392e-07, | |
| "loss": 0.7473, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 1.8510158013544018, | |
| "grad_norm": 3.4391433677619405, | |
| "learning_rate": 3.2685684329588956e-07, | |
| "loss": 0.7304, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.8735891647855532, | |
| "grad_norm": 3.74413308987757, | |
| "learning_rate": 3.1562464364275774e-07, | |
| "loss": 0.7555, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 1.8961625282167043, | |
| "grad_norm": 3.5297868789692677, | |
| "learning_rate": 3.044991270147699e-07, | |
| "loss": 0.7341, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.9187358916478554, | |
| "grad_norm": 3.725919293090452, | |
| "learning_rate": 2.934867308436613e-07, | |
| "loss": 0.7473, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.9413092550790068, | |
| "grad_norm": 3.623859978112656, | |
| "learning_rate": 2.825938271075572e-07, | |
| "loss": 0.7356, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.963882618510158, | |
| "grad_norm": 3.9971274820382057, | |
| "learning_rate": 2.7182671864402856e-07, | |
| "loss": 0.7532, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 1.9864559819413092, | |
| "grad_norm": 3.569291758886745, | |
| "learning_rate": 2.6119163550315194e-07, | |
| "loss": 0.7423, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.0090293453724604, | |
| "grad_norm": 4.160441287080313, | |
| "learning_rate": 2.506947313426854e-07, | |
| "loss": 0.7144, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 2.0316027088036117, | |
| "grad_norm": 3.5037235784557055, | |
| "learning_rate": 2.4034207986744847e-07, | |
| "loss": 0.6957, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.0316027088036117, | |
| "eval_loss": 0.8095582723617554, | |
| "eval_runtime": 109.6376, | |
| "eval_samples_per_second": 57.462, | |
| "eval_steps_per_second": 0.903, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.054176072234763, | |
| "grad_norm": 3.6928343097100265, | |
| "learning_rate": 2.301396713149627e-07, | |
| "loss": 0.6859, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 2.0767494356659144, | |
| "grad_norm": 3.6986435257445645, | |
| "learning_rate": 2.2009340898938738e-07, | |
| "loss": 0.6905, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.0993227990970653, | |
| "grad_norm": 3.4531722388294637, | |
| "learning_rate": 2.1020910584575891e-07, | |
| "loss": 0.6971, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 2.1218961625282167, | |
| "grad_norm": 3.6870711652976436, | |
| "learning_rate": 2.0049248112650563e-07, | |
| "loss": 0.6974, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.144469525959368, | |
| "grad_norm": 3.6781032192277086, | |
| "learning_rate": 1.9094915705218711e-07, | |
| "loss": 0.694, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.1670428893905194, | |
| "grad_norm": 3.7337975408966155, | |
| "learning_rate": 1.8158465556837304e-07, | |
| "loss": 0.701, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.1896162528216703, | |
| "grad_norm": 3.712907514970988, | |
| "learning_rate": 1.7240439515054218e-07, | |
| "loss": 0.6857, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 2.2121896162528216, | |
| "grad_norm": 3.8608434976482684, | |
| "learning_rate": 1.634136876688504e-07, | |
| "loss": 0.6925, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.234762979683973, | |
| "grad_norm": 3.682029857925581, | |
| "learning_rate": 1.5461773531458455e-07, | |
| "loss": 0.7049, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 2.2573363431151243, | |
| "grad_norm": 3.8393609145026484, | |
| "learning_rate": 1.460216275900769e-07, | |
| "loss": 0.7004, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.2573363431151243, | |
| "eval_loss": 0.8108701705932617, | |
| "eval_runtime": 109.7629, | |
| "eval_samples_per_second": 57.396, | |
| "eval_steps_per_second": 0.902, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.2799097065462752, | |
| "grad_norm": 3.7002266455600634, | |
| "learning_rate": 1.3763033836382392e-07, | |
| "loss": 0.691, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 2.3024830699774266, | |
| "grad_norm": 3.792488471905602, | |
| "learning_rate": 1.294487229925132e-07, | |
| "loss": 0.7026, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.325056433408578, | |
| "grad_norm": 3.775955385726932, | |
| "learning_rate": 1.2148151551162345e-07, | |
| "loss": 0.6883, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 2.3476297968397293, | |
| "grad_norm": 3.920690807527362, | |
| "learning_rate": 1.137333258962227e-07, | |
| "loss": 0.7025, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.37020316027088, | |
| "grad_norm": 3.6716732726784973, | |
| "learning_rate": 1.0620863739355135e-07, | |
| "loss": 0.6879, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.3927765237020315, | |
| "grad_norm": 3.8169391966815405, | |
| "learning_rate": 9.891180392893117e-08, | |
| "loss": 0.6968, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.415349887133183, | |
| "grad_norm": 3.8620664380749687, | |
| "learning_rate": 9.184704758650241e-08, | |
| "loss": 0.6914, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 2.4379232505643342, | |
| "grad_norm": 3.767091545661705, | |
| "learning_rate": 8.501845616624798e-08, | |
| "loss": 0.7103, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.460496613995485, | |
| "grad_norm": 3.8929292454919695, | |
| "learning_rate": 7.842998081871493e-08, | |
| "loss": 0.6925, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 2.4830699774266365, | |
| "grad_norm": 3.8571390829470666, | |
| "learning_rate": 7.208543375880594e-08, | |
| "loss": 0.7026, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.4830699774266365, | |
| "eval_loss": 0.8100693821907043, | |
| "eval_runtime": 109.7349, | |
| "eval_samples_per_second": 57.411, | |
| "eval_steps_per_second": 0.902, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.505643340857788, | |
| "grad_norm": 3.7819717082559476, | |
| "learning_rate": 6.598848605996004e-08, | |
| "loss": 0.6829, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 2.528216704288939, | |
| "grad_norm": 3.7991473064242682, | |
| "learning_rate": 6.014266553000074e-08, | |
| "loss": 0.7032, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.55079006772009, | |
| "grad_norm": 3.868317040634209, | |
| "learning_rate": 5.4551354669881145e-08, | |
| "loss": 0.6884, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 2.5733634311512414, | |
| "grad_norm": 3.9808347121085608, | |
| "learning_rate": 4.921778871650539e-08, | |
| "loss": 0.6951, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.595936794582393, | |
| "grad_norm": 3.758581266526973, | |
| "learning_rate": 4.414505377075978e-08, | |
| "loss": 0.6815, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 2.618510158013544, | |
| "grad_norm": 3.543148343804039, | |
| "learning_rate": 3.933608501183788e-08, | |
| "loss": 0.7001, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.6410835214446955, | |
| "grad_norm": 3.88562526533592, | |
| "learning_rate": 3.479366499889058e-08, | |
| "loss": 0.7194, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 2.6636568848758464, | |
| "grad_norm": 3.6861304763463005, | |
| "learning_rate": 3.052042206098537e-08, | |
| "loss": 0.701, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.6862302483069977, | |
| "grad_norm": 3.7669910152435455, | |
| "learning_rate": 2.6518828776306347e-08, | |
| "loss": 0.6901, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 2.708803611738149, | |
| "grad_norm": 3.9930109629654873, | |
| "learning_rate": 2.279120054147393e-08, | |
| "loss": 0.7225, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.708803611738149, | |
| "eval_loss": 0.8094750642776489, | |
| "eval_runtime": 109.8066, | |
| "eval_samples_per_second": 57.374, | |
| "eval_steps_per_second": 0.902, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.7313769751693, | |
| "grad_norm": 3.908911042614429, | |
| "learning_rate": 1.9339694231813252e-08, | |
| "loss": 0.7129, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 2.7539503386004514, | |
| "grad_norm": 3.6295787819862215, | |
| "learning_rate": 1.616630695334592e-08, | |
| "loss": 0.6942, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.7765237020316027, | |
| "grad_norm": 3.7916320186447887, | |
| "learning_rate": 1.3272874887227281e-08, | |
| "loss": 0.7023, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 2.799097065462754, | |
| "grad_norm": 3.819253615915551, | |
| "learning_rate": 1.066107222729712e-08, | |
| "loss": 0.7164, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.8216704288939054, | |
| "grad_norm": 3.7829415672610924, | |
| "learning_rate": 8.332410211360608e-09, | |
| "loss": 0.7038, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 2.8442437923250563, | |
| "grad_norm": 3.9191480311295717, | |
| "learning_rate": 6.288236246757284e-09, | |
| "loss": 0.6883, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.8668171557562077, | |
| "grad_norm": 3.650660911227743, | |
| "learning_rate": 4.529733130726299e-09, | |
| "loss": 0.6832, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 2.889390519187359, | |
| "grad_norm": 3.7262403938852358, | |
| "learning_rate": 3.0579183660177086e-09, | |
| "loss": 0.7142, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.91196388261851, | |
| "grad_norm": 3.6240013513867053, | |
| "learning_rate": 1.8736435721465326e-09, | |
| "loss": 0.6893, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 2.9345372460496613, | |
| "grad_norm": 3.761651981245301, | |
| "learning_rate": 9.775939926296439e-10, | |
| "loss": 0.6961, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.9345372460496613, | |
| "eval_loss": 0.8093209266662598, | |
| "eval_runtime": 109.6169, | |
| "eval_samples_per_second": 57.473, | |
| "eval_steps_per_second": 0.903, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.9571106094808126, | |
| "grad_norm": 3.9259376984897734, | |
| "learning_rate": 3.7028809849098954e-10, | |
| "loss": 0.7092, | |
| "step": 655 | |
| }, | |
| { | |
| "epoch": 2.979683972911964, | |
| "grad_norm": 3.7528491471481313, | |
| "learning_rate": 5.2077288264951166e-11, | |
| "loss": 0.7126, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.9932279909706545, | |
| "step": 663, | |
| "total_flos": 3909114547666944.0, | |
| "train_loss": 0.7950794970108194, | |
| "train_runtime": 10134.0872, | |
| "train_samples_per_second": 16.785, | |
| "train_steps_per_second": 0.065 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 663, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 10000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3909114547666944.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |