| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.7152436298614215, | |
| "eval_steps": 100, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001788109074653554, | |
| "eval_loss": 2.058881998062134, | |
| "eval_runtime": 19.2419, | |
| "eval_samples_per_second": 12.265, | |
| "eval_steps_per_second": 6.132, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.008940545373267769, | |
| "grad_norm": 11.78614616394043, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 7.2335, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.017881090746535537, | |
| "grad_norm": 16.716123580932617, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 6.8512, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02682163611980331, | |
| "grad_norm": 30.8808650970459, | |
| "learning_rate": 5e-05, | |
| "loss": 4.9807, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.035762181493071074, | |
| "grad_norm": 22.647905349731445, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 5.8287, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.044702726866338846, | |
| "grad_norm": 18.210037231445312, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 4.6728, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.05364327223960662, | |
| "grad_norm": 24.852115631103516, | |
| "learning_rate": 0.0001, | |
| "loss": 4.8721, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06258381761287439, | |
| "grad_norm": 21.848241806030273, | |
| "learning_rate": 9.995494831023409e-05, | |
| "loss": 4.8087, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.07152436298614215, | |
| "grad_norm": 38.46466827392578, | |
| "learning_rate": 9.981987442712633e-05, | |
| "loss": 6.3369, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.08046490835940992, | |
| "grad_norm": 40.56308364868164, | |
| "learning_rate": 9.959502176294383e-05, | |
| "loss": 5.1838, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.08940545373267769, | |
| "grad_norm": 50.512901306152344, | |
| "learning_rate": 9.928079551738543e-05, | |
| "loss": 5.9317, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.09834599910594546, | |
| "grad_norm": 10.443129539489746, | |
| "learning_rate": 9.887776194738432e-05, | |
| "loss": 6.75, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.10728654447921324, | |
| "grad_norm": 11.418550491333008, | |
| "learning_rate": 9.838664734667495e-05, | |
| "loss": 6.4133, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.11622708985248101, | |
| "grad_norm": 12.176839828491211, | |
| "learning_rate": 9.780833673696254e-05, | |
| "loss": 4.5906, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.12516763522574878, | |
| "grad_norm": 12.221039772033691, | |
| "learning_rate": 9.714387227305422e-05, | |
| "loss": 4.6752, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.13410818059901655, | |
| "grad_norm": 14.278007507324219, | |
| "learning_rate": 9.639445136482548e-05, | |
| "loss": 4.6537, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.1430487259722843, | |
| "grad_norm": 17.677927017211914, | |
| "learning_rate": 9.55614245194068e-05, | |
| "loss": 4.1464, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.15198927134555207, | |
| "grad_norm": 18.3587589263916, | |
| "learning_rate": 9.464629290747842e-05, | |
| "loss": 4.9328, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.16092981671881984, | |
| "grad_norm": 30.522045135498047, | |
| "learning_rate": 9.365070565805941e-05, | |
| "loss": 5.6559, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1698703620920876, | |
| "grad_norm": 38.24018859863281, | |
| "learning_rate": 9.257645688666556e-05, | |
| "loss": 4.9604, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.17881090746535538, | |
| "grad_norm": 39.934608459472656, | |
| "learning_rate": 9.142548246219212e-05, | |
| "loss": 5.3934, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.17881090746535538, | |
| "eval_loss": 1.4117575883865356, | |
| "eval_runtime": 19.5949, | |
| "eval_samples_per_second": 12.044, | |
| "eval_steps_per_second": 6.022, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.18775145283862316, | |
| "grad_norm": 9.125673294067383, | |
| "learning_rate": 9.019985651834703e-05, | |
| "loss": 6.622, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.19669199821189093, | |
| "grad_norm": 10.328652381896973, | |
| "learning_rate": 8.890178771592199e-05, | |
| "loss": 5.5454, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.2056325435851587, | |
| "grad_norm": 10.82226276397705, | |
| "learning_rate": 8.753361526263621e-05, | |
| "loss": 4.1617, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.21457308895842647, | |
| "grad_norm": 10.679474830627441, | |
| "learning_rate": 8.609780469772623e-05, | |
| "loss": 3.8837, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.22351363433169424, | |
| "grad_norm": 13.074336051940918, | |
| "learning_rate": 8.459694344887732e-05, | |
| "loss": 4.5478, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.23245417970496202, | |
| "grad_norm": 19.095172882080078, | |
| "learning_rate": 8.303373616950408e-05, | |
| "loss": 4.6366, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.24139472507822976, | |
| "grad_norm": 15.73512077331543, | |
| "learning_rate": 8.141099986478212e-05, | |
| "loss": 5.1201, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.25033527045149756, | |
| "grad_norm": 25.20210838317871, | |
| "learning_rate": 7.973165881521434e-05, | |
| "loss": 4.8318, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.25927581582476533, | |
| "grad_norm": 43.75430679321289, | |
| "learning_rate": 7.799873930687978e-05, | |
| "loss": 5.2405, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.2682163611980331, | |
| "grad_norm": 48.83418273925781, | |
| "learning_rate": 7.621536417786159e-05, | |
| "loss": 5.5238, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.2771569065713009, | |
| "grad_norm": 8.100919723510742, | |
| "learning_rate": 7.438474719068173e-05, | |
| "loss": 6.6697, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.2860974519445686, | |
| "grad_norm": 11.477263450622559, | |
| "learning_rate": 7.251018724088367e-05, | |
| "loss": 5.9319, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.29503799731783636, | |
| "grad_norm": 10.063141822814941, | |
| "learning_rate": 7.059506241219965e-05, | |
| "loss": 5.7131, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.30397854269110414, | |
| "grad_norm": 12.019450187683105, | |
| "learning_rate": 6.864282388901544e-05, | |
| "loss": 4.1435, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.3129190880643719, | |
| "grad_norm": 11.132643699645996, | |
| "learning_rate": 6.665698973710288e-05, | |
| "loss": 4.311, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.3218596334376397, | |
| "grad_norm": 16.12037467956543, | |
| "learning_rate": 6.464113856382752e-05, | |
| "loss": 4.0409, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.33080017881090745, | |
| "grad_norm": 14.883207321166992, | |
| "learning_rate": 6.259890306925627e-05, | |
| "loss": 4.0187, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.3397407241841752, | |
| "grad_norm": 19.90201187133789, | |
| "learning_rate": 6.0533963499786314e-05, | |
| "loss": 4.8555, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.348681269557443, | |
| "grad_norm": 18.977224349975586, | |
| "learning_rate": 5.8450041016092464e-05, | |
| "loss": 3.8155, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.35762181493071077, | |
| "grad_norm": 93.56307220458984, | |
| "learning_rate": 5.6350890987343944e-05, | |
| "loss": 6.7862, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.35762181493071077, | |
| "eval_loss": 1.2747399806976318, | |
| "eval_runtime": 19.5936, | |
| "eval_samples_per_second": 12.045, | |
| "eval_steps_per_second": 6.022, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.36656236030397854, | |
| "grad_norm": 10.30494213104248, | |
| "learning_rate": 5.4240296223775465e-05, | |
| "loss": 6.2965, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.3755029056772463, | |
| "grad_norm": 9.586296081542969, | |
| "learning_rate": 5.212206015980742e-05, | |
| "loss": 5.2106, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.3844434510505141, | |
| "grad_norm": 11.460832595825195, | |
| "learning_rate": 5e-05, | |
| "loss": 5.0099, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.39338399642378186, | |
| "grad_norm": 15.051645278930664, | |
| "learning_rate": 4.78779398401926e-05, | |
| "loss": 4.0896, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.40232454179704963, | |
| "grad_norm": 15.79628849029541, | |
| "learning_rate": 4.575970377622456e-05, | |
| "loss": 3.9889, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.4112650871703174, | |
| "grad_norm": 9.98121166229248, | |
| "learning_rate": 4.364910901265606e-05, | |
| "loss": 4.068, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.4202056325435852, | |
| "grad_norm": 13.449023246765137, | |
| "learning_rate": 4.1549958983907555e-05, | |
| "loss": 4.0041, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.42914617791685294, | |
| "grad_norm": 17.6621150970459, | |
| "learning_rate": 3.94660365002137e-05, | |
| "loss": 4.5437, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.4380867232901207, | |
| "grad_norm": 42.47966384887695, | |
| "learning_rate": 3.740109693074375e-05, | |
| "loss": 4.8385, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.4470272686633885, | |
| "grad_norm": 54.10409164428711, | |
| "learning_rate": 3.5358861436172485e-05, | |
| "loss": 5.6786, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.45596781403665626, | |
| "grad_norm": 8.634622573852539, | |
| "learning_rate": 3.334301026289712e-05, | |
| "loss": 6.4977, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.46490835940992403, | |
| "grad_norm": 11.310672760009766, | |
| "learning_rate": 3.135717611098458e-05, | |
| "loss": 4.4356, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.47384890478319175, | |
| "grad_norm": 11.240716934204102, | |
| "learning_rate": 2.9404937587800375e-05, | |
| "loss": 4.3816, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.4827894501564595, | |
| "grad_norm": 10.130899429321289, | |
| "learning_rate": 2.748981275911633e-05, | |
| "loss": 4.2432, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.4917299955297273, | |
| "grad_norm": 11.680331230163574, | |
| "learning_rate": 2.5615252809318284e-05, | |
| "loss": 3.5449, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.5006705409029951, | |
| "grad_norm": 14.900591850280762, | |
| "learning_rate": 2.3784635822138424e-05, | |
| "loss": 4.2528, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.5096110862762628, | |
| "grad_norm": 17.08873748779297, | |
| "learning_rate": 2.2001260693120233e-05, | |
| "loss": 4.8141, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.5185516316495307, | |
| "grad_norm": 24.609628677368164, | |
| "learning_rate": 2.026834118478567e-05, | |
| "loss": 3.6354, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.5274921770227984, | |
| "grad_norm": 29.8326358795166, | |
| "learning_rate": 1.858900013521788e-05, | |
| "loss": 4.305, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.5364327223960662, | |
| "grad_norm": 40.15150833129883, | |
| "learning_rate": 1.6966263830495936e-05, | |
| "loss": 4.9746, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5364327223960662, | |
| "eval_loss": 1.2247991561889648, | |
| "eval_runtime": 19.5982, | |
| "eval_samples_per_second": 12.042, | |
| "eval_steps_per_second": 6.021, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5453732677693339, | |
| "grad_norm": 6.730564117431641, | |
| "learning_rate": 1.5403056551122697e-05, | |
| "loss": 5.6797, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.5543138131426018, | |
| "grad_norm": 12.260231971740723, | |
| "learning_rate": 1.3902195302273779e-05, | |
| "loss": 4.9479, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.5632543585158695, | |
| "grad_norm": 12.601056098937988, | |
| "learning_rate": 1.246638473736378e-05, | |
| "loss": 4.2043, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.5721949038891372, | |
| "grad_norm": 12.597325325012207, | |
| "learning_rate": 1.1098212284078036e-05, | |
| "loss": 4.047, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.581135449262405, | |
| "grad_norm": 12.159523010253906, | |
| "learning_rate": 9.800143481652979e-06, | |
| "loss": 4.2291, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.5900759946356727, | |
| "grad_norm": 12.575312614440918, | |
| "learning_rate": 8.574517537807897e-06, | |
| "loss": 3.4932, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.5990165400089406, | |
| "grad_norm": 25.68616485595703, | |
| "learning_rate": 7.423543113334436e-06, | |
| "loss": 4.4752, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.6079570853822083, | |
| "grad_norm": 19.73110008239746, | |
| "learning_rate": 6.349294341940593e-06, | |
| "loss": 4.4499, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.6168976307554761, | |
| "grad_norm": 33.66725158691406, | |
| "learning_rate": 5.353707092521582e-06, | |
| "loss": 3.9091, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 0.6258381761287438, | |
| "grad_norm": 37.051727294921875, | |
| "learning_rate": 4.43857548059321e-06, | |
| "loss": 5.7529, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.6347787215020116, | |
| "grad_norm": 11.532958030700684, | |
| "learning_rate": 3.605548635174533e-06, | |
| "loss": 5.8252, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 0.6437192668752794, | |
| "grad_norm": 10.616311073303223, | |
| "learning_rate": 2.85612772694579e-06, | |
| "loss": 4.048, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.6526598122485472, | |
| "grad_norm": 9.950262069702148, | |
| "learning_rate": 2.191663263037458e-06, | |
| "loss": 4.3658, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 0.6616003576218149, | |
| "grad_norm": 9.809283256530762, | |
| "learning_rate": 1.6133526533250565e-06, | |
| "loss": 3.7123, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.6705409029950827, | |
| "grad_norm": 8.985926628112793, | |
| "learning_rate": 1.1222380526156928e-06, | |
| "loss": 3.5311, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.6794814483683504, | |
| "grad_norm": 15.09157657623291, | |
| "learning_rate": 7.192044826145771e-07, | |
| "loss": 3.9273, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.6884219937416183, | |
| "grad_norm": 19.525718688964844, | |
| "learning_rate": 4.049782370561583e-07, | |
| "loss": 4.0169, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 0.697362539114886, | |
| "grad_norm": 22.638031005859375, | |
| "learning_rate": 1.8012557287367392e-07, | |
| "loss": 4.7521, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.7063030844881538, | |
| "grad_norm": 29.394636154174805, | |
| "learning_rate": 4.5051689765929214e-08, | |
| "loss": 4.773, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 0.7152436298614215, | |
| "grad_norm": 39.56925964355469, | |
| "learning_rate": 0.0, | |
| "loss": 4.2009, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7152436298614215, | |
| "eval_loss": 1.1803392171859741, | |
| "eval_runtime": 19.5987, | |
| "eval_samples_per_second": 12.042, | |
| "eval_steps_per_second": 6.021, | |
| "step": 400 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 400, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.4091489359010202e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |