| { | |
| "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": 17.11013535387158, | |
| "learning_rate": 5e-07, | |
| "loss": 1.7379, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.045146726862302484, | |
| "grad_norm": 12.959848430726268, | |
| "learning_rate": 1e-06, | |
| "loss": 1.601, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06772009029345373, | |
| "grad_norm": 7.576528956673404, | |
| "learning_rate": 9.998553453302385e-07, | |
| "loss": 1.2543, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.09029345372460497, | |
| "grad_norm": 4.2302392012204235, | |
| "learning_rate": 9.99421465020848e-07, | |
| "loss": 1.1527, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.11286681715575621, | |
| "grad_norm": 3.847937840469522, | |
| "learning_rate": 9.9869861012308e-07, | |
| "loss": 1.0663, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.13544018058690746, | |
| "grad_norm": 3.6316845940431746, | |
| "learning_rate": 9.976871988942804e-07, | |
| "loss": 1.0105, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1580135440180587, | |
| "grad_norm": 3.649401460423432, | |
| "learning_rate": 9.963878165558785e-07, | |
| "loss": 0.9909, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.18058690744920994, | |
| "grad_norm": 4.05306836191529, | |
| "learning_rate": 9.948012149547666e-07, | |
| "loss": 0.9578, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.20316027088036118, | |
| "grad_norm": 3.6396577845814337, | |
| "learning_rate": 9.929283121282675e-07, | |
| "loss": 0.9684, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.22573363431151242, | |
| "grad_norm": 3.8537199144308762, | |
| "learning_rate": 9.9077019177294e-07, | |
| "loss": 0.9416, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.22573363431151242, | |
| "eval_loss": 0.9273125529289246, | |
| "eval_runtime": 109.7674, | |
| "eval_samples_per_second": 57.394, | |
| "eval_steps_per_second": 0.902, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.24830699774266365, | |
| "grad_norm": 3.3681487821748397, | |
| "learning_rate": 9.88328102617534e-07, | |
| "loss": 0.9352, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.2708803611738149, | |
| "grad_norm": 3.4337087526601975, | |
| "learning_rate": 9.856034577004504e-07, | |
| "loss": 0.9378, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.29345372460496616, | |
| "grad_norm": 3.4486053479519154, | |
| "learning_rate": 9.82597833552132e-07, | |
| "loss": 0.9416, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.3160270880361174, | |
| "grad_norm": 3.238782431149946, | |
| "learning_rate": 9.793129692828533e-07, | |
| "loss": 0.9265, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.33860045146726864, | |
| "grad_norm": 3.2810204129248435, | |
| "learning_rate": 9.757507655764384e-07, | |
| "loss": 0.9034, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.3611738148984199, | |
| "grad_norm": 3.398462798378792, | |
| "learning_rate": 9.719132835904906e-07, | |
| "loss": 0.8817, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.3837471783295711, | |
| "grad_norm": 3.6031011626579232, | |
| "learning_rate": 9.678027437637677e-07, | |
| "loss": 0.8923, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.40632054176072235, | |
| "grad_norm": 3.4992072745130063, | |
| "learning_rate": 9.634215245313939e-07, | |
| "loss": 0.8776, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4288939051918736, | |
| "grad_norm": 3.4568242774115876, | |
| "learning_rate": 9.587721609486543e-07, | |
| "loss": 0.8738, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.45146726862302483, | |
| "grad_norm": 3.347304172280455, | |
| "learning_rate": 9.538573432241637e-07, | |
| "loss": 0.876, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.45146726862302483, | |
| "eval_loss": 0.8719745874404907, | |
| "eval_runtime": 109.9293, | |
| "eval_samples_per_second": 57.31, | |
| "eval_steps_per_second": 0.901, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.47404063205417607, | |
| "grad_norm": 3.442543320251002, | |
| "learning_rate": 9.486799151632612e-07, | |
| "loss": 0.8803, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.4966139954853273, | |
| "grad_norm": 3.5980312527331395, | |
| "learning_rate": 9.432428725225326e-07, | |
| "loss": 0.8753, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.5191873589164786, | |
| "grad_norm": 3.675561348684254, | |
| "learning_rate": 9.375493612764085e-07, | |
| "loss": 0.8855, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.5417607223476298, | |
| "grad_norm": 3.3697248672451634, | |
| "learning_rate": 9.316026757968454e-07, | |
| "loss": 0.8625, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.5643340857787811, | |
| "grad_norm": 3.473011914846237, | |
| "learning_rate": 9.2540625694714e-07, | |
| "loss": 0.872, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.5869074492099323, | |
| "grad_norm": 3.4789702681993337, | |
| "learning_rate": 9.189636900909817e-07, | |
| "loss": 0.8585, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.6094808126410836, | |
| "grad_norm": 3.4492734978467374, | |
| "learning_rate": 9.122787030178949e-07, | |
| "loss": 0.8446, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.6320541760722348, | |
| "grad_norm": 3.525194093646181, | |
| "learning_rate": 9.053551637862692e-07, | |
| "loss": 0.8554, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.654627539503386, | |
| "grad_norm": 3.5743743068358746, | |
| "learning_rate": 8.98197078485229e-07, | |
| "loss": 0.8609, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.6772009029345373, | |
| "grad_norm": 3.398056434064616, | |
| "learning_rate": 8.908085889166357e-07, | |
| "loss": 0.8328, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.6772009029345373, | |
| "eval_loss": 0.847376823425293, | |
| "eval_runtime": 109.9357, | |
| "eval_samples_per_second": 57.306, | |
| "eval_steps_per_second": 0.901, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.6997742663656885, | |
| "grad_norm": 3.5580216310657655, | |
| "learning_rate": 8.831939701985636e-07, | |
| "loss": 0.848, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.7223476297968398, | |
| "grad_norm": 3.736752885842491, | |
| "learning_rate": 8.75357628291637e-07, | |
| "loss": 0.8429, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.744920993227991, | |
| "grad_norm": 3.5381861103919916, | |
| "learning_rate": 8.673040974496584e-07, | |
| "loss": 0.8445, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.7674943566591422, | |
| "grad_norm": 3.638564106955557, | |
| "learning_rate": 8.590380375960053e-07, | |
| "loss": 0.8345, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.7900677200902935, | |
| "grad_norm": 3.3891030445610255, | |
| "learning_rate": 8.505642316273111e-07, | |
| "loss": 0.8471, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.8126410835214447, | |
| "grad_norm": 3.5786902411065604, | |
| "learning_rate": 8.418875826459919e-07, | |
| "loss": 0.8572, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.835214446952596, | |
| "grad_norm": 3.6449275586594134, | |
| "learning_rate": 8.330131111232201e-07, | |
| "loss": 0.8321, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.8577878103837472, | |
| "grad_norm": 3.1629627713809536, | |
| "learning_rate": 8.239459519939851e-07, | |
| "loss": 0.816, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.8803611738148984, | |
| "grad_norm": 3.4582550141931416, | |
| "learning_rate": 8.14691351685925e-07, | |
| "loss": 0.8303, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.9029345372460497, | |
| "grad_norm": 3.5511684536445864, | |
| "learning_rate": 8.052546650836453e-07, | |
| "loss": 0.8288, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.9029345372460497, | |
| "eval_loss": 0.8302638530731201, | |
| "eval_runtime": 109.5812, | |
| "eval_samples_per_second": 57.492, | |
| "eval_steps_per_second": 0.903, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.9255079006772009, | |
| "grad_norm": 3.5003235834203665, | |
| "learning_rate": 7.956413524302823e-07, | |
| "loss": 0.8243, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.9480812641083521, | |
| "grad_norm": 3.555478472903463, | |
| "learning_rate": 7.858569761681047e-07, | |
| "loss": 0.8176, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.9706546275395034, | |
| "grad_norm": 3.5362559032917775, | |
| "learning_rate": 7.759071977199806e-07, | |
| "loss": 0.8151, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.9932279909706546, | |
| "grad_norm": 3.2803583751821126, | |
| "learning_rate": 7.657977742135725e-07, | |
| "loss": 0.8251, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.0158013544018059, | |
| "grad_norm": 3.433858537440282, | |
| "learning_rate": 7.555345551501557e-07, | |
| "loss": 0.799, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.0383747178329572, | |
| "grad_norm": 3.532208184837943, | |
| "learning_rate": 7.451234790199871e-07, | |
| "loss": 0.7549, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.0609480812641083, | |
| "grad_norm": 4.248609105994399, | |
| "learning_rate": 7.345705698661852e-07, | |
| "loss": 0.7943, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.0835214446952597, | |
| "grad_norm": 3.631037157637433, | |
| "learning_rate": 7.23881933799104e-07, | |
| "loss": 0.7706, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.1060948081264108, | |
| "grad_norm": 3.666559132397504, | |
| "learning_rate": 7.130637554632257e-07, | |
| "loss": 0.7681, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.1286681715575622, | |
| "grad_norm": 3.973093725177555, | |
| "learning_rate": 7.021222944586088e-07, | |
| "loss": 0.772, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.1286681715575622, | |
| "eval_loss": 0.8229889273643494, | |
| "eval_runtime": 109.7153, | |
| "eval_samples_per_second": 57.421, | |
| "eval_steps_per_second": 0.902, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.1512415349887133, | |
| "grad_norm": 3.3931909367135784, | |
| "learning_rate": 6.910638817189664e-07, | |
| "loss": 0.7549, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.1738148984198646, | |
| "grad_norm": 3.3943674694946413, | |
| "learning_rate": 6.798949158484705e-07, | |
| "loss": 0.7699, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.1963882618510158, | |
| "grad_norm": 3.600159247119083, | |
| "learning_rate": 6.686218594193993e-07, | |
| "loss": 0.7621, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.2189616252821671, | |
| "grad_norm": 3.5847356389330005, | |
| "learning_rate": 6.572512352327726e-07, | |
| "loss": 0.7686, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.2415349887133182, | |
| "grad_norm": 3.5603747270255126, | |
| "learning_rate": 6.457896225441371e-07, | |
| "loss": 0.7601, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.2641083521444696, | |
| "grad_norm": 3.546694170007199, | |
| "learning_rate": 6.342436532566865e-07, | |
| "loss": 0.7723, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.2866817155756207, | |
| "grad_norm": 3.560362460851058, | |
| "learning_rate": 6.226200080839182e-07, | |
| "loss": 0.7474, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.309255079006772, | |
| "grad_norm": 3.60150415758672, | |
| "learning_rate": 6.109254126840479e-07, | |
| "loss": 0.7694, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.3318284424379232, | |
| "grad_norm": 3.5529975551326864, | |
| "learning_rate": 5.991666337684176e-07, | |
| "loss": 0.7511, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.3544018058690745, | |
| "grad_norm": 3.4780226362354103, | |
| "learning_rate": 5.873504751861507e-07, | |
| "loss": 0.7434, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.3544018058690745, | |
| "eval_loss": 0.8149687051773071, | |
| "eval_runtime": 109.9888, | |
| "eval_samples_per_second": 57.279, | |
| "eval_steps_per_second": 0.9, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.3769751693002257, | |
| "grad_norm": 3.7930095963516948, | |
| "learning_rate": 5.754837739873178e-07, | |
| "loss": 0.7391, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 1.399548532731377, | |
| "grad_norm": 3.6471827689760214, | |
| "learning_rate": 5.635733964668909e-07, | |
| "loss": 0.7601, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.4221218961625282, | |
| "grad_norm": 3.618221803291648, | |
| "learning_rate": 5.516262341917778e-07, | |
| "loss": 0.7571, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 1.4446952595936795, | |
| "grad_norm": 3.5821066578321052, | |
| "learning_rate": 5.396492000132325e-07, | |
| "loss": 0.7562, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.4672686230248306, | |
| "grad_norm": 3.487176364922937, | |
| "learning_rate": 5.276492240669503e-07, | |
| "loss": 0.7545, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.489841986455982, | |
| "grad_norm": 3.576844855283387, | |
| "learning_rate": 5.156332497631621e-07, | |
| "loss": 0.7527, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.5124153498871333, | |
| "grad_norm": 3.9241166204798317, | |
| "learning_rate": 5.036082297690464e-07, | |
| "loss": 0.74, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 1.5349887133182845, | |
| "grad_norm": 3.719210568412769, | |
| "learning_rate": 4.915811219857882e-07, | |
| "loss": 0.7644, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.5575620767494356, | |
| "grad_norm": 3.8943797624821244, | |
| "learning_rate": 4.795588855226055e-07, | |
| "loss": 0.7681, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 1.580135440180587, | |
| "grad_norm": 3.5819561294880695, | |
| "learning_rate": 4.6754847667008004e-07, | |
| "loss": 0.7387, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.580135440180587, | |
| "eval_loss": 0.8084157109260559, | |
| "eval_runtime": 109.8341, | |
| "eval_samples_per_second": 57.359, | |
| "eval_steps_per_second": 0.901, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.6027088036117383, | |
| "grad_norm": 4.171144264405167, | |
| "learning_rate": 4.5555684487511693e-07, | |
| "loss": 0.7434, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 1.6252821670428894, | |
| "grad_norm": 3.392679445857191, | |
| "learning_rate": 4.435909287198646e-07, | |
| "loss": 0.7553, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.6478555304740405, | |
| "grad_norm": 3.705567890275386, | |
| "learning_rate": 4.316576519069226e-07, | |
| "loss": 0.763, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 1.670428893905192, | |
| "grad_norm": 3.402848569658215, | |
| "learning_rate": 4.197639192531573e-07, | |
| "loss": 0.7619, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.6930022573363432, | |
| "grad_norm": 3.7305097686546147, | |
| "learning_rate": 4.079166126944453e-07, | |
| "loss": 0.7465, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.7155756207674944, | |
| "grad_norm": 3.5614904023987175, | |
| "learning_rate": 3.9612258730365823e-07, | |
| "loss": 0.7564, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.7381489841986455, | |
| "grad_norm": 3.662383997694909, | |
| "learning_rate": 3.843886673241883e-07, | |
| "loss": 0.7525, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 1.7607223476297968, | |
| "grad_norm": 3.5121198800285907, | |
| "learning_rate": 3.7272164222131387e-07, | |
| "loss": 0.7664, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.7832957110609482, | |
| "grad_norm": 3.6927580052581717, | |
| "learning_rate": 3.611282627536887e-07, | |
| "loss": 0.7378, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 1.8058690744920993, | |
| "grad_norm": 3.6445526526932546, | |
| "learning_rate": 3.496152370672255e-07, | |
| "loss": 0.7409, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.8058690744920993, | |
| "eval_loss": 0.8030996322631836, | |
| "eval_runtime": 109.7877, | |
| "eval_samples_per_second": 57.383, | |
| "eval_steps_per_second": 0.902, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.8284424379232505, | |
| "grad_norm": 3.5221955021123406, | |
| "learning_rate": 3.381892268136392e-07, | |
| "loss": 0.7582, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 1.8510158013544018, | |
| "grad_norm": 3.5764273204454717, | |
| "learning_rate": 3.2685684329588956e-07, | |
| "loss": 0.7442, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.8735891647855532, | |
| "grad_norm": 3.54563665313358, | |
| "learning_rate": 3.1562464364275774e-07, | |
| "loss": 0.7437, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 1.8961625282167043, | |
| "grad_norm": 3.6949864387414535, | |
| "learning_rate": 3.044991270147699e-07, | |
| "loss": 0.7423, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.9187358916478554, | |
| "grad_norm": 3.675638978380515, | |
| "learning_rate": 2.934867308436613e-07, | |
| "loss": 0.7443, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.9413092550790068, | |
| "grad_norm": 3.6497669806576294, | |
| "learning_rate": 2.825938271075572e-07, | |
| "loss": 0.7475, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.963882618510158, | |
| "grad_norm": 3.8644787157192915, | |
| "learning_rate": 2.7182671864402856e-07, | |
| "loss": 0.7371, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 1.9864559819413092, | |
| "grad_norm": 3.64079890206456, | |
| "learning_rate": 2.6119163550315194e-07, | |
| "loss": 0.7427, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.0090293453724604, | |
| "grad_norm": 4.007442882651605, | |
| "learning_rate": 2.506947313426854e-07, | |
| "loss": 0.7306, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 2.0316027088036117, | |
| "grad_norm": 3.607359710331693, | |
| "learning_rate": 2.4034207986744847e-07, | |
| "loss": 0.7112, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.0316027088036117, | |
| "eval_loss": 0.8032650947570801, | |
| "eval_runtime": 109.6605, | |
| "eval_samples_per_second": 57.45, | |
| "eval_steps_per_second": 0.903, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.054176072234763, | |
| "grad_norm": 3.9730315312501507, | |
| "learning_rate": 2.301396713149627e-07, | |
| "loss": 0.7181, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 2.0767494356659144, | |
| "grad_norm": 3.8749839105157635, | |
| "learning_rate": 2.2009340898938738e-07, | |
| "loss": 0.6868, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.0993227990970653, | |
| "grad_norm": 3.617704984713086, | |
| "learning_rate": 2.1020910584575891e-07, | |
| "loss": 0.6879, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 2.1218961625282167, | |
| "grad_norm": 3.52665889407724, | |
| "learning_rate": 2.0049248112650563e-07, | |
| "loss": 0.6966, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.144469525959368, | |
| "grad_norm": 3.693088533913318, | |
| "learning_rate": 1.9094915705218711e-07, | |
| "loss": 0.6954, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.1670428893905194, | |
| "grad_norm": 3.798502667433261, | |
| "learning_rate": 1.8158465556837304e-07, | |
| "loss": 0.7154, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.1896162528216703, | |
| "grad_norm": 3.809227909402352, | |
| "learning_rate": 1.7240439515054218e-07, | |
| "loss": 0.6915, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 2.2121896162528216, | |
| "grad_norm": 3.964820136012577, | |
| "learning_rate": 1.634136876688504e-07, | |
| "loss": 0.6997, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.234762979683973, | |
| "grad_norm": 3.5741262004158068, | |
| "learning_rate": 1.5461773531458455e-07, | |
| "loss": 0.6868, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 2.2573363431151243, | |
| "grad_norm": 3.9054213147478376, | |
| "learning_rate": 1.460216275900769e-07, | |
| "loss": 0.7123, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.2573363431151243, | |
| "eval_loss": 0.8042254447937012, | |
| "eval_runtime": 109.7878, | |
| "eval_samples_per_second": 57.383, | |
| "eval_steps_per_second": 0.902, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.2799097065462752, | |
| "grad_norm": 3.8693292107518573, | |
| "learning_rate": 1.3763033836382392e-07, | |
| "loss": 0.7066, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 2.3024830699774266, | |
| "grad_norm": 3.941292944572669, | |
| "learning_rate": 1.294487229925132e-07, | |
| "loss": 0.711, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.325056433408578, | |
| "grad_norm": 3.6787665693791496, | |
| "learning_rate": 1.2148151551162345e-07, | |
| "loss": 0.6717, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 2.3476297968397293, | |
| "grad_norm": 3.86722467174776, | |
| "learning_rate": 1.137333258962227e-07, | |
| "loss": 0.6835, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.37020316027088, | |
| "grad_norm": 3.958290971612969, | |
| "learning_rate": 1.0620863739355135e-07, | |
| "loss": 0.7041, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.3927765237020315, | |
| "grad_norm": 4.1390165215470365, | |
| "learning_rate": 9.891180392893117e-08, | |
| "loss": 0.6952, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.415349887133183, | |
| "grad_norm": 3.6879955563550184, | |
| "learning_rate": 9.184704758650241e-08, | |
| "loss": 0.7108, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 2.4379232505643342, | |
| "grad_norm": 3.8009234892174444, | |
| "learning_rate": 8.501845616624798e-08, | |
| "loss": 0.6977, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.460496613995485, | |
| "grad_norm": 4.1238817767449705, | |
| "learning_rate": 7.842998081871493e-08, | |
| "loss": 0.696, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 2.4830699774266365, | |
| "grad_norm": 3.7983544943293426, | |
| "learning_rate": 7.208543375880594e-08, | |
| "loss": 0.7057, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.4830699774266365, | |
| "eval_loss": 0.8041459918022156, | |
| "eval_runtime": 109.7979, | |
| "eval_samples_per_second": 57.378, | |
| "eval_steps_per_second": 0.902, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.505643340857788, | |
| "grad_norm": 3.8588530963050798, | |
| "learning_rate": 6.598848605996004e-08, | |
| "loss": 0.6895, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 2.528216704288939, | |
| "grad_norm": 3.9145683115902945, | |
| "learning_rate": 6.014266553000074e-08, | |
| "loss": 0.7101, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.55079006772009, | |
| "grad_norm": 3.75474980179681, | |
| "learning_rate": 5.4551354669881145e-08, | |
| "loss": 0.6801, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 2.5733634311512414, | |
| "grad_norm": 3.7778184283211234, | |
| "learning_rate": 4.921778871650539e-08, | |
| "loss": 0.7006, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.595936794582393, | |
| "grad_norm": 3.7245539109756773, | |
| "learning_rate": 4.414505377075978e-08, | |
| "loss": 0.6836, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 2.618510158013544, | |
| "grad_norm": 3.8925958915786145, | |
| "learning_rate": 3.933608501183788e-08, | |
| "loss": 0.6926, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.6410835214446955, | |
| "grad_norm": 3.8795075554625926, | |
| "learning_rate": 3.479366499889058e-08, | |
| "loss": 0.6928, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 2.6636568848758464, | |
| "grad_norm": 3.8858578602563463, | |
| "learning_rate": 3.052042206098537e-08, | |
| "loss": 0.6768, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.6862302483069977, | |
| "grad_norm": 3.7825570201880647, | |
| "learning_rate": 2.6518828776306347e-08, | |
| "loss": 0.7098, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 2.708803611738149, | |
| "grad_norm": 4.01533662282639, | |
| "learning_rate": 2.279120054147393e-08, | |
| "loss": 0.6976, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.708803611738149, | |
| "eval_loss": 0.8028527498245239, | |
| "eval_runtime": 109.7212, | |
| "eval_samples_per_second": 57.418, | |
| "eval_steps_per_second": 0.902, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.7313769751693, | |
| "grad_norm": 4.018506896194764, | |
| "learning_rate": 1.9339694231813252e-08, | |
| "loss": 0.7002, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 2.7539503386004514, | |
| "grad_norm": 3.7530557112622973, | |
| "learning_rate": 1.616630695334592e-08, | |
| "loss": 0.7035, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.7765237020316027, | |
| "grad_norm": 3.725603099520227, | |
| "learning_rate": 1.3272874887227281e-08, | |
| "loss": 0.6868, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 2.799097065462754, | |
| "grad_norm": 3.8271764781590165, | |
| "learning_rate": 1.066107222729712e-08, | |
| "loss": 0.692, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.8216704288939054, | |
| "grad_norm": 3.7012687609338784, | |
| "learning_rate": 8.332410211360608e-09, | |
| "loss": 0.7102, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 2.8442437923250563, | |
| "grad_norm": 3.8420799277743045, | |
| "learning_rate": 6.288236246757284e-09, | |
| "loss": 0.7124, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.8668171557562077, | |
| "grad_norm": 4.073031095233782, | |
| "learning_rate": 4.529733130726299e-09, | |
| "loss": 0.6918, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 2.889390519187359, | |
| "grad_norm": 3.9624071770097204, | |
| "learning_rate": 3.0579183660177086e-09, | |
| "loss": 0.7078, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.91196388261851, | |
| "grad_norm": 3.7493043258434504, | |
| "learning_rate": 1.8736435721465326e-09, | |
| "loss": 0.6908, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 2.9345372460496613, | |
| "grad_norm": 3.7396790845979346, | |
| "learning_rate": 9.775939926296439e-10, | |
| "loss": 0.6986, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.9345372460496613, | |
| "eval_loss": 0.802590548992157, | |
| "eval_runtime": 109.7038, | |
| "eval_samples_per_second": 57.427, | |
| "eval_steps_per_second": 0.902, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.9571106094808126, | |
| "grad_norm": 3.5845554984233927, | |
| "learning_rate": 3.7028809849098954e-10, | |
| "loss": 0.6925, | |
| "step": 655 | |
| }, | |
| { | |
| "epoch": 2.979683972911964, | |
| "grad_norm": 3.787744121195186, | |
| "learning_rate": 5.2077288264951166e-11, | |
| "loss": 0.703, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.9932279909706545, | |
| "step": 663, | |
| "total_flos": 3909114547666944.0, | |
| "train_loss": 0.7953385170408683, | |
| "train_runtime": 10139.3898, | |
| "train_samples_per_second": 16.776, | |
| "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 | |
| } | |