| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9878213802435725, | |
| "eval_steps": 50, | |
| "global_step": 552, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02706359945872801, | |
| "grad_norm": 18.62096133756944, | |
| "learning_rate": 5e-07, | |
| "loss": 1.7202, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.05412719891745602, | |
| "grad_norm": 12.425223769043265, | |
| "learning_rate": 1e-06, | |
| "loss": 1.5858, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08119079837618404, | |
| "grad_norm": 7.4903509661138825, | |
| "learning_rate": 9.997900331216397e-07, | |
| "loss": 1.2962, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.10825439783491204, | |
| "grad_norm": 4.423017017969494, | |
| "learning_rate": 9.991603088309193e-07, | |
| "loss": 1.1481, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.13531799729364005, | |
| "grad_norm": 4.05084233893573, | |
| "learning_rate": 9.981113560128126e-07, | |
| "loss": 1.0683, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.16238159675236807, | |
| "grad_norm": 3.6533701555355194, | |
| "learning_rate": 9.966440556487147e-07, | |
| "loss": 1.0347, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.18944519621109607, | |
| "grad_norm": 3.5509774718765037, | |
| "learning_rate": 9.947596400765342e-07, | |
| "loss": 0.9829, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.2165087956698241, | |
| "grad_norm": 3.389187665280499, | |
| "learning_rate": 9.924596919556916e-07, | |
| "loss": 0.9751, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2435723951285521, | |
| "grad_norm": 3.7276246548668697, | |
| "learning_rate": 9.897461429378964e-07, | |
| "loss": 0.9534, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2706359945872801, | |
| "grad_norm": 3.6238376734485906, | |
| "learning_rate": 9.866212720448147e-07, | |
| "loss": 0.9375, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2706359945872801, | |
| "eval_loss": 0.9354712963104248, | |
| "eval_runtime": 91.9855, | |
| "eval_samples_per_second": 57.074, | |
| "eval_steps_per_second": 0.902, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2976995940460081, | |
| "grad_norm": 3.650038732579689, | |
| "learning_rate": 9.830877037539935e-07, | |
| "loss": 0.9241, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.32476319350473615, | |
| "grad_norm": 3.282303443355455, | |
| "learning_rate": 9.791484057946465e-07, | |
| "loss": 0.9301, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.35182679296346414, | |
| "grad_norm": 3.5367161380439316, | |
| "learning_rate": 9.748066866551555e-07, | |
| "loss": 0.9141, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.37889039242219213, | |
| "grad_norm": 3.384471001472233, | |
| "learning_rate": 9.700661928043785e-07, | |
| "loss": 0.9099, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.4059539918809202, | |
| "grad_norm": 3.6506849242542603, | |
| "learning_rate": 9.649309056290984e-07, | |
| "loss": 0.9034, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.4330175913396482, | |
| "grad_norm": 3.5334381489912654, | |
| "learning_rate": 9.594051380901858e-07, | |
| "loss": 0.8755, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.46008119079837617, | |
| "grad_norm": 3.617605597301526, | |
| "learning_rate": 9.534935311002834e-07, | |
| "loss": 0.887, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.4871447902571042, | |
| "grad_norm": 3.8303351034768016, | |
| "learning_rate": 9.472010496260544e-07, | |
| "loss": 0.8868, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5142083897158322, | |
| "grad_norm": 3.4133330247065485, | |
| "learning_rate": 9.405329785182678e-07, | |
| "loss": 0.8685, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.5412719891745602, | |
| "grad_norm": 3.598327334615042, | |
| "learning_rate": 9.334949180732244e-07, | |
| "loss": 0.8824, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5412719891745602, | |
| "eval_loss": 0.8775722980499268, | |
| "eval_runtime": 91.8061, | |
| "eval_samples_per_second": 57.186, | |
| "eval_steps_per_second": 0.904, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5683355886332883, | |
| "grad_norm": 3.4406636804059803, | |
| "learning_rate": 9.260927793292497e-07, | |
| "loss": 0.8847, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.5953991880920162, | |
| "grad_norm": 3.7093344605385363, | |
| "learning_rate": 9.183327791022047e-07, | |
| "loss": 0.8716, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.6224627875507442, | |
| "grad_norm": 3.6150174879821275, | |
| "learning_rate": 9.102214347641843e-07, | |
| "loss": 0.879, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.6495263870094723, | |
| "grad_norm": 3.323783724540155, | |
| "learning_rate": 9.017655587697883e-07, | |
| "loss": 0.849, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.6765899864682002, | |
| "grad_norm": 3.4870778817811705, | |
| "learning_rate": 8.929722529345623e-07, | |
| "loss": 0.8623, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.7036535859269283, | |
| "grad_norm": 3.3887325632646554, | |
| "learning_rate": 8.83848902470413e-07, | |
| "loss": 0.8821, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.7307171853856563, | |
| "grad_norm": 3.4877253492258253, | |
| "learning_rate": 8.744031697830088e-07, | |
| "loss": 0.864, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.7577807848443843, | |
| "grad_norm": 3.4539745143868643, | |
| "learning_rate": 8.646429880363746e-07, | |
| "loss": 0.8469, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.7848443843031123, | |
| "grad_norm": 3.6501092648768645, | |
| "learning_rate": 8.545765544900846e-07, | |
| "loss": 0.8488, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.8119079837618404, | |
| "grad_norm": 3.6661416233299233, | |
| "learning_rate": 8.442123236146508e-07, | |
| "loss": 0.8549, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.8119079837618404, | |
| "eval_loss": 0.8518306612968445, | |
| "eval_runtime": 91.7277, | |
| "eval_samples_per_second": 57.235, | |
| "eval_steps_per_second": 0.905, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.8389715832205683, | |
| "grad_norm": 3.6360925835102895, | |
| "learning_rate": 8.33558999990887e-07, | |
| "loss": 0.846, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.8660351826792964, | |
| "grad_norm": 3.325785655756218, | |
| "learning_rate": 8.22625530999215e-07, | |
| "loss": 0.8485, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.8930987821380244, | |
| "grad_norm": 3.5191969682264492, | |
| "learning_rate": 8.114210993050502e-07, | |
| "loss": 0.86, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.9201623815967523, | |
| "grad_norm": 3.505912957460407, | |
| "learning_rate": 7.999551151465791e-07, | |
| "loss": 0.8445, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.9472259810554804, | |
| "grad_norm": 3.567280090097035, | |
| "learning_rate": 7.88237208431406e-07, | |
| "loss": 0.8366, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.9742895805142084, | |
| "grad_norm": 3.5124657783307143, | |
| "learning_rate": 7.762772206487065e-07, | |
| "loss": 0.8377, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.0013531799729365, | |
| "grad_norm": 3.661655068528439, | |
| "learning_rate": 7.640851966036805e-07, | |
| "loss": 0.8402, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.0284167794316643, | |
| "grad_norm": 3.443540295008513, | |
| "learning_rate": 7.516713759812464e-07, | |
| "loss": 0.7742, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.0554803788903924, | |
| "grad_norm": 3.5551780337145975, | |
| "learning_rate": 7.390461847460628e-07, | |
| "loss": 0.7723, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 1.0825439783491204, | |
| "grad_norm": 3.4346090387850965, | |
| "learning_rate": 7.262202263860988e-07, | |
| "loss": 0.7825, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0825439783491204, | |
| "eval_loss": 0.8397604823112488, | |
| "eval_runtime": 91.7114, | |
| "eval_samples_per_second": 57.245, | |
| "eval_steps_per_second": 0.905, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.1096075778078485, | |
| "grad_norm": 3.346387935116934, | |
| "learning_rate": 7.1320427300711e-07, | |
| "loss": 0.7622, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 1.1366711772665765, | |
| "grad_norm": 3.6975628986759532, | |
| "learning_rate": 7.000092562854959e-07, | |
| "loss": 0.7683, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.1637347767253043, | |
| "grad_norm": 3.534180683955908, | |
| "learning_rate": 6.866462582871401e-07, | |
| "loss": 0.7536, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.1907983761840324, | |
| "grad_norm": 3.6284911794109322, | |
| "learning_rate": 6.731265021599436e-07, | |
| "loss": 0.775, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.2178619756427604, | |
| "grad_norm": 3.9686622547711816, | |
| "learning_rate": 6.594613427078674e-07, | |
| "loss": 0.7736, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.2449255751014885, | |
| "grad_norm": 3.4953313507774797, | |
| "learning_rate": 6.456622568544011e-07, | |
| "loss": 0.7723, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.2719891745602165, | |
| "grad_norm": 3.600480261317538, | |
| "learning_rate": 6.317408340034684e-07, | |
| "loss": 0.7675, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.2990527740189446, | |
| "grad_norm": 4.160648083143722, | |
| "learning_rate": 6.177087663058625e-07, | |
| "loss": 0.786, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.3261163734776726, | |
| "grad_norm": 3.554038996352176, | |
| "learning_rate": 6.035778388393893e-07, | |
| "loss": 0.7569, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.3531799729364005, | |
| "grad_norm": 3.4314153682677517, | |
| "learning_rate": 5.893599197109624e-07, | |
| "loss": 0.759, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.3531799729364005, | |
| "eval_loss": 0.8301268219947815, | |
| "eval_runtime": 91.7237, | |
| "eval_samples_per_second": 57.237, | |
| "eval_steps_per_second": 0.905, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.3802435723951285, | |
| "grad_norm": 3.405464319479644, | |
| "learning_rate": 5.750669500889666e-07, | |
| "loss": 0.7549, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.4073071718538566, | |
| "grad_norm": 3.500530938219073, | |
| "learning_rate": 5.607109341742578e-07, | |
| "loss": 0.7801, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.4343707713125846, | |
| "grad_norm": 3.726591852861668, | |
| "learning_rate": 5.463039291182256e-07, | |
| "loss": 0.7765, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.4614343707713127, | |
| "grad_norm": 3.619160960040315, | |
| "learning_rate": 5.318580348963825e-07, | |
| "loss": 0.7662, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.4884979702300405, | |
| "grad_norm": 3.4244376461150208, | |
| "learning_rate": 5.173853841459877e-07, | |
| "loss": 0.7554, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.5155615696887685, | |
| "grad_norm": 3.700667266849164, | |
| "learning_rate": 5.028981319762399e-07, | |
| "loss": 0.7616, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.5426251691474966, | |
| "grad_norm": 3.567507416375757, | |
| "learning_rate": 4.884084457595956e-07, | |
| "loss": 0.7818, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.5696887686062246, | |
| "grad_norm": 3.7802418009731755, | |
| "learning_rate": 4.7392849491278817e-07, | |
| "loss": 0.7691, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.5967523680649527, | |
| "grad_norm": 3.501336855850476, | |
| "learning_rate": 4.5947044067613e-07, | |
| "loss": 0.7681, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.6238159675236807, | |
| "grad_norm": 3.656073742916776, | |
| "learning_rate": 4.4504642589968217e-07, | |
| "loss": 0.7676, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.6238159675236807, | |
| "eval_loss": 0.8219158053398132, | |
| "eval_runtime": 91.7533, | |
| "eval_samples_per_second": 57.219, | |
| "eval_steps_per_second": 0.905, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.6508795669824088, | |
| "grad_norm": 3.704023712304986, | |
| "learning_rate": 4.3066856484486847e-07, | |
| "loss": 0.7602, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 1.6779431664411368, | |
| "grad_norm": 3.504247839905232, | |
| "learning_rate": 4.1634893301010165e-07, | |
| "loss": 0.7589, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.7050067658998647, | |
| "grad_norm": 3.470803136242184, | |
| "learning_rate": 4.0209955698896445e-07, | |
| "loss": 0.7475, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 1.7320703653585927, | |
| "grad_norm": 3.5356035894949045, | |
| "learning_rate": 3.8793240436946385e-07, | |
| "loss": 0.7548, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.7591339648173205, | |
| "grad_norm": 3.559368281490328, | |
| "learning_rate": 3.738593736828426e-07, | |
| "loss": 0.761, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.7861975642760486, | |
| "grad_norm": 3.4859137753013942, | |
| "learning_rate": 3.598922844103902e-07, | |
| "loss": 0.7704, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.8132611637347766, | |
| "grad_norm": 3.6109852467133607, | |
| "learning_rate": 3.4604286705664397e-07, | |
| "loss": 0.7578, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 1.8403247631935047, | |
| "grad_norm": 3.478328670865475, | |
| "learning_rate": 3.323227532973193e-07, | |
| "loss": 0.7548, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.8673883626522327, | |
| "grad_norm": 3.670794273772087, | |
| "learning_rate": 3.187434662102434e-07, | |
| "loss": 0.7624, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 1.8944519621109608, | |
| "grad_norm": 3.6564249976504963, | |
| "learning_rate": 3.0531641059749634e-07, | |
| "loss": 0.7624, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.8944519621109608, | |
| "eval_loss": 0.81601482629776, | |
| "eval_runtime": 91.595, | |
| "eval_samples_per_second": 57.318, | |
| "eval_steps_per_second": 0.906, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.9215155615696888, | |
| "grad_norm": 3.8249796571875754, | |
| "learning_rate": 2.920528634068885e-07, | |
| "loss": 0.7342, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 1.9485791610284169, | |
| "grad_norm": 3.4390639442266044, | |
| "learning_rate": 2.789639642608184e-07, | |
| "loss": 0.7552, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.975642760487145, | |
| "grad_norm": 3.681741592514357, | |
| "learning_rate": 2.6606070610046526e-07, | |
| "loss": 0.7578, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 2.002706359945873, | |
| "grad_norm": 3.7841567374808815, | |
| "learning_rate": 2.533539259531757e-07, | |
| "loss": 0.755, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.029769959404601, | |
| "grad_norm": 3.406665416810156, | |
| "learning_rate": 2.408542958307957e-07, | |
| "loss": 0.6915, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.0568335588633286, | |
| "grad_norm": 3.627816770705192, | |
| "learning_rate": 2.2857231376659514e-07, | |
| "loss": 0.7139, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.0838971583220567, | |
| "grad_norm": 3.8009146897382164, | |
| "learning_rate": 2.1651829499831043e-07, | |
| "loss": 0.7127, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 2.1109607577807847, | |
| "grad_norm": 3.7612594070053578, | |
| "learning_rate": 2.0470236330471126e-07, | |
| "loss": 0.7188, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.138024357239513, | |
| "grad_norm": 3.881855102735637, | |
| "learning_rate": 1.9313444250296846e-07, | |
| "loss": 0.7319, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 2.165087956698241, | |
| "grad_norm": 3.8666379554848103, | |
| "learning_rate": 1.818242481139613e-07, | |
| "loss": 0.6976, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.165087956698241, | |
| "eval_loss": 0.8192901611328125, | |
| "eval_runtime": 91.7299, | |
| "eval_samples_per_second": 57.233, | |
| "eval_steps_per_second": 0.905, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.192151556156969, | |
| "grad_norm": 3.6983015115543676, | |
| "learning_rate": 1.7078127920252783e-07, | |
| "loss": 0.7018, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 2.219215155615697, | |
| "grad_norm": 3.4856114871878408, | |
| "learning_rate": 1.600148103995087e-07, | |
| "loss": 0.7062, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.246278755074425, | |
| "grad_norm": 3.7609793781320042, | |
| "learning_rate": 1.4953388411228602e-07, | |
| "loss": 0.714, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 2.273342354533153, | |
| "grad_norm": 3.7207726096154206, | |
| "learning_rate": 1.3934730293035936e-07, | |
| "loss": 0.7028, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.300405953991881, | |
| "grad_norm": 3.9694406933094193, | |
| "learning_rate": 1.2946362223233614e-07, | |
| "loss": 0.7023, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.3274695534506087, | |
| "grad_norm": 3.63014175931209, | |
| "learning_rate": 1.198911430005478e-07, | |
| "loss": 0.7022, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.3545331529093367, | |
| "grad_norm": 3.7202986392499295, | |
| "learning_rate": 1.1063790484932462e-07, | |
| "loss": 0.7083, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 2.381596752368065, | |
| "grad_norm": 3.640272659998316, | |
| "learning_rate": 1.0171167927278368e-07, | |
| "loss": 0.6996, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.408660351826793, | |
| "grad_norm": 3.830774844166055, | |
| "learning_rate": 9.311996311780446e-08, | |
| "loss": 0.7032, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 2.435723951285521, | |
| "grad_norm": 3.835735833278635, | |
| "learning_rate": 8.486997228767012e-08, | |
| "loss": 0.7004, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.435723951285521, | |
| "eval_loss": 0.8185598254203796, | |
| "eval_runtime": 91.6401, | |
| "eval_samples_per_second": 57.289, | |
| "eval_steps_per_second": 0.906, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.462787550744249, | |
| "grad_norm": 3.8096679910803526, | |
| "learning_rate": 7.696863568166518e-08, | |
| "loss": 0.6951, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 2.489851150202977, | |
| "grad_norm": 3.9855447041279612, | |
| "learning_rate": 6.942258937571771e-08, | |
| "loss": 0.7148, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.516914749661705, | |
| "grad_norm": 3.8539380052102943, | |
| "learning_rate": 6.2238171048975e-08, | |
| "loss": 0.7011, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 2.543978349120433, | |
| "grad_norm": 4.0430762185988725, | |
| "learning_rate": 5.5421414660992705e-08, | |
| "loss": 0.7206, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.571041948579161, | |
| "grad_norm": 3.936740213024643, | |
| "learning_rate": 4.8978045384008125e-08, | |
| "loss": 0.7116, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.598105548037889, | |
| "grad_norm": 3.9575285108660196, | |
| "learning_rate": 4.2913474794554036e-08, | |
| "loss": 0.7192, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.6251691474966172, | |
| "grad_norm": 3.8004579862768604, | |
| "learning_rate": 3.723279632845155e-08, | |
| "loss": 0.7213, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 2.6522327469553453, | |
| "grad_norm": 3.9896579923804953, | |
| "learning_rate": 3.194078100299863e-08, | |
| "loss": 0.7058, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.6792963464140733, | |
| "grad_norm": 3.485486386140113, | |
| "learning_rate": 2.7041873409947734e-08, | |
| "loss": 0.7155, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 2.706359945872801, | |
| "grad_norm": 3.7310098177240625, | |
| "learning_rate": 2.2540187982637627e-08, | |
| "loss": 0.7133, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.706359945872801, | |
| "eval_loss": 0.817565381526947, | |
| "eval_runtime": 91.7069, | |
| "eval_samples_per_second": 57.248, | |
| "eval_steps_per_second": 0.905, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.733423545331529, | |
| "grad_norm": 3.6063602953215654, | |
| "learning_rate": 1.8439505540414458e-08, | |
| "loss": 0.7024, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 2.760487144790257, | |
| "grad_norm": 3.6598065079223527, | |
| "learning_rate": 1.4743270113244277e-08, | |
| "loss": 0.7039, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.787550744248985, | |
| "grad_norm": 3.7941708315282847, | |
| "learning_rate": 1.1454586049184589e-08, | |
| "loss": 0.7096, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 2.814614343707713, | |
| "grad_norm": 4.0192290237097765, | |
| "learning_rate": 8.576215407142651e-09, | |
| "loss": 0.6872, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.841677943166441, | |
| "grad_norm": 3.671284433438612, | |
| "learning_rate": 6.110575637112425e-09, | |
| "loss": 0.6934, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.8687415426251692, | |
| "grad_norm": 4.26951192895122, | |
| "learning_rate": 4.059737549836517e-09, | |
| "loss": 0.7066, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.8958051420838973, | |
| "grad_norm": 3.732791917417492, | |
| "learning_rate": 2.425423577599783e-09, | |
| "loss": 0.6894, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 2.9228687415426253, | |
| "grad_norm": 3.626360935172459, | |
| "learning_rate": 1.209006327614226e-09, | |
| "loss": 0.7122, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.949932341001353, | |
| "grad_norm": 3.649121404665413, | |
| "learning_rate": 4.115074292109777e-10, | |
| "loss": 0.7, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 2.976995940460081, | |
| "grad_norm": 3.550586809689184, | |
| "learning_rate": 3.3596675806824013e-11, | |
| "loss": 0.7, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.976995940460081, | |
| "eval_loss": 0.8177570104598999, | |
| "eval_runtime": 91.665, | |
| "eval_samples_per_second": 57.274, | |
| "eval_steps_per_second": 0.905, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.9878213802435725, | |
| "step": 552, | |
| "total_flos": 3254608239525888.0, | |
| "train_loss": 0.8081599063631417, | |
| "train_runtime": 8273.4491, | |
| "train_samples_per_second": 17.133, | |
| "train_steps_per_second": 0.067 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 552, | |
| "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": 3254608239525888.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |