smolm-autoreg-bpe-counterfactual_babylm_aann_high_variability_numeral-seed_211-1e-3
/
trainer_state.json
| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 371920, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.8909159302711487, | |
| "learning_rate": 3.125e-05, | |
| "loss": 6.228, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 1.025515079498291, | |
| "learning_rate": 6.25e-05, | |
| "loss": 5.0018, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.9241665005683899, | |
| "learning_rate": 9.375e-05, | |
| "loss": 4.6753, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.8754993677139282, | |
| "learning_rate": 0.000125, | |
| "loss": 4.4594, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.7558003664016724, | |
| "learning_rate": 0.00015625, | |
| "loss": 4.3007, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.7471924424171448, | |
| "learning_rate": 0.0001875, | |
| "loss": 4.1771, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.6875430345535278, | |
| "learning_rate": 0.00021875, | |
| "loss": 4.0791, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.6728542447090149, | |
| "learning_rate": 0.00025, | |
| "loss": 3.9784, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.6585622429847717, | |
| "learning_rate": 0.00028125000000000003, | |
| "loss": 3.9146, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.5960397124290466, | |
| "learning_rate": 0.00031246875000000003, | |
| "loss": 3.847, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.576011061668396, | |
| "learning_rate": 0.00034368749999999997, | |
| "loss": 3.8063, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.5716710686683655, | |
| "learning_rate": 0.0003749375, | |
| "loss": 3.7672, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.5205326080322266, | |
| "learning_rate": 0.00040615625, | |
| "loss": 3.7304, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.47286731004714966, | |
| "learning_rate": 0.00043740625, | |
| "loss": 3.6967, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.4275617301464081, | |
| "learning_rate": 0.00046865625, | |
| "loss": 3.6717, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.4085034430027008, | |
| "learning_rate": 0.000499875, | |
| "loss": 3.6497, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.40013808012008667, | |
| "learning_rate": 0.00053109375, | |
| "loss": 3.6225, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.3701670169830322, | |
| "learning_rate": 0.00056234375, | |
| "loss": 3.61, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.35835879185489444, | |
| "eval_loss": 3.795727014541626, | |
| "eval_runtime": 153.3517, | |
| "eval_samples_per_second": 377.674, | |
| "eval_steps_per_second": 5.901, | |
| "step": 18596 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 0.34899449348449707, | |
| "learning_rate": 0.00059359375, | |
| "loss": 3.579, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 0.37780293822288513, | |
| "learning_rate": 0.0006248437500000001, | |
| "loss": 3.5464, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "grad_norm": 0.3226105272769928, | |
| "learning_rate": 0.0006560625, | |
| "loss": 3.5387, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 0.3026217222213745, | |
| "learning_rate": 0.0006873125, | |
| "loss": 3.5284, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 0.2845563590526581, | |
| "learning_rate": 0.0007185625, | |
| "loss": 3.5209, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "grad_norm": 0.28052306175231934, | |
| "learning_rate": 0.0007498125, | |
| "loss": 3.5094, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "grad_norm": 0.34453320503234863, | |
| "learning_rate": 0.0007810312499999999, | |
| "loss": 3.4934, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.30819350481033325, | |
| "learning_rate": 0.00081228125, | |
| "loss": 3.4931, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 0.30646008253097534, | |
| "learning_rate": 0.0008435000000000001, | |
| "loss": 3.477, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "grad_norm": 0.24530355632305145, | |
| "learning_rate": 0.0008747500000000001, | |
| "loss": 3.4684, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 0.287979394197464, | |
| "learning_rate": 0.0009059375, | |
| "loss": 3.4633, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "grad_norm": 0.2318340390920639, | |
| "learning_rate": 0.0009371875, | |
| "loss": 3.4523, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "grad_norm": 0.21640178561210632, | |
| "learning_rate": 0.0009684375, | |
| "loss": 3.4511, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 0.23932267725467682, | |
| "learning_rate": 0.0009996562500000001, | |
| "loss": 3.44, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 0.23123665153980255, | |
| "learning_rate": 0.0009970904918804424, | |
| "loss": 3.4308, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "grad_norm": 0.22664424777030945, | |
| "learning_rate": 0.0009941486232054601, | |
| "loss": 3.4153, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 0.21310342848300934, | |
| "learning_rate": 0.0009912067545304777, | |
| "loss": 3.4106, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "grad_norm": 0.2162679135799408, | |
| "learning_rate": 0.0009882678277241704, | |
| "loss": 3.4031, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "grad_norm": 0.21360158920288086, | |
| "learning_rate": 0.0009853259590491882, | |
| "loss": 3.3861, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.38014852732029175, | |
| "eval_loss": 3.564697742462158, | |
| "eval_runtime": 154.5153, | |
| "eval_samples_per_second": 374.83, | |
| "eval_steps_per_second": 5.857, | |
| "step": 37192 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 0.2235065996646881, | |
| "learning_rate": 0.0009823899741115557, | |
| "loss": 3.3397, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 0.2304013967514038, | |
| "learning_rate": 0.0009794481054365733, | |
| "loss": 3.3289, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 0.22438882291316986, | |
| "learning_rate": 0.000976506236761591, | |
| "loss": 3.326, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 0.21580879390239716, | |
| "learning_rate": 0.0009735643680866086, | |
| "loss": 3.3208, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "grad_norm": 0.21414010226726532, | |
| "learning_rate": 0.0009706283831489763, | |
| "loss": 3.3202, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "grad_norm": 0.20920276641845703, | |
| "learning_rate": 0.0009676865144739939, | |
| "loss": 3.3111, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "grad_norm": 0.19549404084682465, | |
| "learning_rate": 0.0009647446457990115, | |
| "loss": 3.308, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "grad_norm": 0.2047666609287262, | |
| "learning_rate": 0.0009618027771240292, | |
| "loss": 3.307, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "grad_norm": 0.19435334205627441, | |
| "learning_rate": 0.0009588609084490469, | |
| "loss": 3.3007, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "grad_norm": 0.20967498421669006, | |
| "learning_rate": 0.0009559190397740644, | |
| "loss": 3.293, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "grad_norm": 0.22619305551052094, | |
| "learning_rate": 0.0009529801129677571, | |
| "loss": 3.2911, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "grad_norm": 0.1997978389263153, | |
| "learning_rate": 0.0009500382442927748, | |
| "loss": 3.286, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "grad_norm": 0.22306489944458008, | |
| "learning_rate": 0.0009470963756177925, | |
| "loss": 3.2839, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "grad_norm": 0.20853149890899658, | |
| "learning_rate": 0.000944157448811485, | |
| "loss": 3.2745, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.20844997465610504, | |
| "learning_rate": 0.0009412155801365028, | |
| "loss": 3.2657, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "grad_norm": 0.19329632818698883, | |
| "learning_rate": 0.0009382766533301954, | |
| "loss": 3.2683, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 0.23765866458415985, | |
| "learning_rate": 0.000935334784655213, | |
| "loss": 3.2637, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 0.2281341701745987, | |
| "learning_rate": 0.0009323929159802307, | |
| "loss": 3.2591, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.39145331347307755, | |
| "eval_loss": 3.4933156967163086, | |
| "eval_runtime": 154.9004, | |
| "eval_samples_per_second": 373.898, | |
| "eval_steps_per_second": 5.842, | |
| "step": 55788 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "grad_norm": 0.2204052358865738, | |
| "learning_rate": 0.0009294539891739233, | |
| "loss": 3.2431, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "grad_norm": 0.2415602058172226, | |
| "learning_rate": 0.000926512120498941, | |
| "loss": 3.1919, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 0.19874005019664764, | |
| "learning_rate": 0.0009235702518239586, | |
| "loss": 3.196, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "grad_norm": 0.1932218074798584, | |
| "learning_rate": 0.0009206283831489762, | |
| "loss": 3.1988, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "grad_norm": 0.21343468129634857, | |
| "learning_rate": 0.0009176894563426689, | |
| "loss": 3.1961, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "grad_norm": 0.2427150458097458, | |
| "learning_rate": 0.0009147475876676865, | |
| "loss": 3.1967, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "grad_norm": 0.24537405371665955, | |
| "learning_rate": 0.0009118086608613791, | |
| "loss": 3.1992, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "grad_norm": 0.20690293610095978, | |
| "learning_rate": 0.0009088667921863968, | |
| "loss": 3.1979, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "grad_norm": 0.25561264157295227, | |
| "learning_rate": 0.0009059249235114144, | |
| "loss": 3.1956, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "grad_norm": 0.20544248819351196, | |
| "learning_rate": 0.000902985996705107, | |
| "loss": 3.1969, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "grad_norm": 0.19939689338207245, | |
| "learning_rate": 0.0009000441280301248, | |
| "loss": 3.1913, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 0.2552586495876312, | |
| "learning_rate": 0.0008971052012238174, | |
| "loss": 3.1915, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "grad_norm": 0.23190921545028687, | |
| "learning_rate": 0.000894163332548835, | |
| "loss": 3.1966, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "grad_norm": 0.24762412905693054, | |
| "learning_rate": 0.0008912244057425277, | |
| "loss": 3.1867, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "grad_norm": 0.2145838439464569, | |
| "learning_rate": 0.0008882825370675454, | |
| "loss": 3.1905, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "grad_norm": 0.2245662808418274, | |
| "learning_rate": 0.0008853436102612379, | |
| "loss": 3.1846, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "grad_norm": 0.2066437155008316, | |
| "learning_rate": 0.0008824017415862556, | |
| "loss": 3.1832, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "grad_norm": 0.21306444704532623, | |
| "learning_rate": 0.0008794628147799483, | |
| "loss": 3.1834, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "grad_norm": 0.22630509734153748, | |
| "learning_rate": 0.000876520946104966, | |
| "loss": 3.1836, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.3972446414495344, | |
| "eval_loss": 3.4326329231262207, | |
| "eval_runtime": 154.0547, | |
| "eval_samples_per_second": 375.951, | |
| "eval_steps_per_second": 5.875, | |
| "step": 74384 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "grad_norm": 0.19157204031944275, | |
| "learning_rate": 0.0008735790774299835, | |
| "loss": 3.1363, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "grad_norm": 0.20826587080955505, | |
| "learning_rate": 0.0008706401506236762, | |
| "loss": 3.1164, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "grad_norm": 0.20335273444652557, | |
| "learning_rate": 0.0008676982819486939, | |
| "loss": 3.119, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "grad_norm": 0.20953166484832764, | |
| "learning_rate": 0.0008647564132737114, | |
| "loss": 3.1245, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "grad_norm": 0.20731399953365326, | |
| "learning_rate": 0.0008618145445987291, | |
| "loss": 3.1238, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "grad_norm": 0.23300333321094513, | |
| "learning_rate": 0.0008588756177924218, | |
| "loss": 3.1252, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "grad_norm": 0.20891304314136505, | |
| "learning_rate": 0.0008559337491174393, | |
| "loss": 3.1288, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "grad_norm": 0.21623115241527557, | |
| "learning_rate": 0.000852994822311132, | |
| "loss": 3.1285, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "grad_norm": 0.2474987506866455, | |
| "learning_rate": 0.0008500529536361497, | |
| "loss": 3.1262, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "grad_norm": 0.2118910700082779, | |
| "learning_rate": 0.0008471140268298423, | |
| "loss": 3.127, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "grad_norm": 0.24061526358127594, | |
| "learning_rate": 0.00084417215815486, | |
| "loss": 3.131, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "grad_norm": 0.22474774718284607, | |
| "learning_rate": 0.0008412302894798777, | |
| "loss": 3.1276, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "grad_norm": 0.19421574473381042, | |
| "learning_rate": 0.0008382884208048953, | |
| "loss": 3.127, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "grad_norm": 0.20476099848747253, | |
| "learning_rate": 0.0008353494939985879, | |
| "loss": 3.128, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "grad_norm": 0.23917576670646667, | |
| "learning_rate": 0.0008324076253236056, | |
| "loss": 3.1285, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "grad_norm": 0.213016539812088, | |
| "learning_rate": 0.0008294657566486232, | |
| "loss": 3.1281, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "grad_norm": 0.20960000157356262, | |
| "learning_rate": 0.0008265238879736408, | |
| "loss": 3.1314, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "grad_norm": 0.1963956207036972, | |
| "learning_rate": 0.0008235849611673335, | |
| "loss": 3.1278, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.40181854228669783, | |
| "eval_loss": 3.4132988452911377, | |
| "eval_runtime": 154.1416, | |
| "eval_samples_per_second": 375.739, | |
| "eval_steps_per_second": 5.871, | |
| "step": 92980 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.26301759481430054, | |
| "learning_rate": 0.0008206430924923511, | |
| "loss": 3.121, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "grad_norm": 0.21968483924865723, | |
| "learning_rate": 0.0008177012238173688, | |
| "loss": 3.0579, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "grad_norm": 0.23824404180049896, | |
| "learning_rate": 0.0008147622970110614, | |
| "loss": 3.0621, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "grad_norm": 0.20740585029125214, | |
| "learning_rate": 0.000811820428336079, | |
| "loss": 3.0653, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "grad_norm": 0.21885009109973907, | |
| "learning_rate": 0.0008088785596610967, | |
| "loss": 3.0721, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "grad_norm": 0.27921879291534424, | |
| "learning_rate": 0.0008059425747234644, | |
| "loss": 3.0724, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "grad_norm": 0.2494860142469406, | |
| "learning_rate": 0.0008030007060484819, | |
| "loss": 3.0771, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "grad_norm": 0.2149578034877777, | |
| "learning_rate": 0.0008000588373734997, | |
| "loss": 3.0778, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "grad_norm": 0.24940645694732666, | |
| "learning_rate": 0.0007971169686985174, | |
| "loss": 3.079, | |
| "step": 101000 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "grad_norm": 0.2192409187555313, | |
| "learning_rate": 0.00079417804189221, | |
| "loss": 3.0803, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "grad_norm": 0.21941109001636505, | |
| "learning_rate": 0.0007912361732172276, | |
| "loss": 3.0771, | |
| "step": 103000 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "grad_norm": 0.21313978731632233, | |
| "learning_rate": 0.0007882943045422453, | |
| "loss": 3.0784, | |
| "step": 104000 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "grad_norm": 0.27852264046669006, | |
| "learning_rate": 0.0007853553777359379, | |
| "loss": 3.0842, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "grad_norm": 0.27452683448791504, | |
| "learning_rate": 0.0007824164509296305, | |
| "loss": 3.0822, | |
| "step": 106000 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "grad_norm": 0.22761721909046173, | |
| "learning_rate": 0.0007794745822546482, | |
| "loss": 3.0824, | |
| "step": 107000 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "grad_norm": 0.21608448028564453, | |
| "learning_rate": 0.0007765356554483409, | |
| "loss": 3.0859, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "grad_norm": 0.26219895482063293, | |
| "learning_rate": 0.0007735937867733584, | |
| "loss": 3.0817, | |
| "step": 109000 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "grad_norm": 0.2277887463569641, | |
| "learning_rate": 0.0007706519180983761, | |
| "loss": 3.0799, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "grad_norm": 0.20768152177333832, | |
| "learning_rate": 0.0007677100494233938, | |
| "loss": 3.0795, | |
| "step": 111000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.40326990139989005, | |
| "eval_loss": 3.4124019145965576, | |
| "eval_runtime": 154.0264, | |
| "eval_samples_per_second": 376.02, | |
| "eval_steps_per_second": 5.876, | |
| "step": 111576 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "grad_norm": 0.20890597999095917, | |
| "learning_rate": 0.0007647681807484113, | |
| "loss": 3.049, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "grad_norm": 0.23515057563781738, | |
| "learning_rate": 0.000761829253942104, | |
| "loss": 3.0134, | |
| "step": 113000 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "grad_norm": 0.23229765892028809, | |
| "learning_rate": 0.0007588873852671217, | |
| "loss": 3.0212, | |
| "step": 114000 | |
| }, | |
| { | |
| "epoch": 6.18, | |
| "grad_norm": 0.26751187443733215, | |
| "learning_rate": 0.0007559455165921393, | |
| "loss": 3.0246, | |
| "step": 115000 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.25669464468955994, | |
| "learning_rate": 0.0007530095316545069, | |
| "loss": 3.0349, | |
| "step": 116000 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "grad_norm": 0.2544041872024536, | |
| "learning_rate": 0.0007500676629795246, | |
| "loss": 3.0371, | |
| "step": 117000 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "grad_norm": 0.23453743755817413, | |
| "learning_rate": 0.0007471257943045423, | |
| "loss": 3.0346, | |
| "step": 118000 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "grad_norm": 0.2783219814300537, | |
| "learning_rate": 0.0007441839256295599, | |
| "loss": 3.0379, | |
| "step": 119000 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "grad_norm": 0.24288877844810486, | |
| "learning_rate": 0.0007412420569545776, | |
| "loss": 3.0353, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "grad_norm": 0.21807478368282318, | |
| "learning_rate": 0.0007383031301482702, | |
| "loss": 3.0422, | |
| "step": 121000 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "grad_norm": 0.2313011735677719, | |
| "learning_rate": 0.0007353642033419628, | |
| "loss": 3.0417, | |
| "step": 122000 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "grad_norm": 0.23170366883277893, | |
| "learning_rate": 0.0007324223346669805, | |
| "loss": 3.0423, | |
| "step": 123000 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "grad_norm": 0.21878008544445038, | |
| "learning_rate": 0.0007294804659919982, | |
| "loss": 3.0422, | |
| "step": 124000 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.2428036332130432, | |
| "learning_rate": 0.0007265385973170157, | |
| "loss": 3.0479, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "grad_norm": 0.22587792575359344, | |
| "learning_rate": 0.0007235967286420334, | |
| "loss": 3.0425, | |
| "step": 126000 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "grad_norm": 0.22501161694526672, | |
| "learning_rate": 0.000720657801835726, | |
| "loss": 3.0446, | |
| "step": 127000 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "grad_norm": 0.23849289119243622, | |
| "learning_rate": 0.0007177159331607437, | |
| "loss": 3.0415, | |
| "step": 128000 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "grad_norm": 0.23556004464626312, | |
| "learning_rate": 0.0007147770063544363, | |
| "loss": 3.0478, | |
| "step": 129000 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "grad_norm": 0.22832949459552765, | |
| "learning_rate": 0.000711835137679454, | |
| "loss": 3.0444, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.4052325248213447, | |
| "eval_loss": 3.3837218284606934, | |
| "eval_runtime": 154.0952, | |
| "eval_samples_per_second": 375.852, | |
| "eval_steps_per_second": 5.873, | |
| "step": 130172 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "grad_norm": 0.21719269454479218, | |
| "learning_rate": 0.0007088932690044717, | |
| "loss": 2.9885, | |
| "step": 131000 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "grad_norm": 0.24770499765872955, | |
| "learning_rate": 0.0007059543421981642, | |
| "loss": 2.9813, | |
| "step": 132000 | |
| }, | |
| { | |
| "epoch": 7.15, | |
| "grad_norm": 0.23300251364707947, | |
| "learning_rate": 0.000703012473523182, | |
| "loss": 2.9884, | |
| "step": 133000 | |
| }, | |
| { | |
| "epoch": 7.21, | |
| "grad_norm": 0.2238881140947342, | |
| "learning_rate": 0.0007000735467168746, | |
| "loss": 2.9932, | |
| "step": 134000 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "grad_norm": 0.22716933488845825, | |
| "learning_rate": 0.0006971316780418923, | |
| "loss": 2.9953, | |
| "step": 135000 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "grad_norm": 0.2647615075111389, | |
| "learning_rate": 0.0006941898093669099, | |
| "loss": 3.0028, | |
| "step": 136000 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "grad_norm": 0.2218490093946457, | |
| "learning_rate": 0.0006912508825606025, | |
| "loss": 3.0034, | |
| "step": 137000 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "grad_norm": 0.20896905660629272, | |
| "learning_rate": 0.0006883090138856202, | |
| "loss": 3.0015, | |
| "step": 138000 | |
| }, | |
| { | |
| "epoch": 7.47, | |
| "grad_norm": 0.24288326501846313, | |
| "learning_rate": 0.0006853700870793128, | |
| "loss": 3.0057, | |
| "step": 139000 | |
| }, | |
| { | |
| "epoch": 7.53, | |
| "grad_norm": 0.23336203396320343, | |
| "learning_rate": 0.0006824282184043304, | |
| "loss": 3.0104, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 7.58, | |
| "grad_norm": 0.22705671191215515, | |
| "learning_rate": 0.0006794863497293481, | |
| "loss": 3.0079, | |
| "step": 141000 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "grad_norm": 0.24445344507694244, | |
| "learning_rate": 0.0006765474229230408, | |
| "loss": 3.0095, | |
| "step": 142000 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "grad_norm": 0.24330651760101318, | |
| "learning_rate": 0.0006736055542480583, | |
| "loss": 3.0088, | |
| "step": 143000 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "grad_norm": 0.23043465614318848, | |
| "learning_rate": 0.000670666627441751, | |
| "loss": 3.0093, | |
| "step": 144000 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "grad_norm": 0.21775782108306885, | |
| "learning_rate": 0.0006677247587667687, | |
| "loss": 3.0126, | |
| "step": 145000 | |
| }, | |
| { | |
| "epoch": 7.85, | |
| "grad_norm": 0.2638954222202301, | |
| "learning_rate": 0.0006647858319604612, | |
| "loss": 3.014, | |
| "step": 146000 | |
| }, | |
| { | |
| "epoch": 7.9, | |
| "grad_norm": 0.24298764765262604, | |
| "learning_rate": 0.0006618439632854789, | |
| "loss": 3.0144, | |
| "step": 147000 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "grad_norm": 0.2365158647298813, | |
| "learning_rate": 0.0006589020946104966, | |
| "loss": 3.0115, | |
| "step": 148000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.4068079041596424, | |
| "eval_loss": 3.366936683654785, | |
| "eval_runtime": 153.7426, | |
| "eval_samples_per_second": 376.714, | |
| "eval_steps_per_second": 5.886, | |
| "step": 148768 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "grad_norm": 0.2332482486963272, | |
| "learning_rate": 0.0006559631678041892, | |
| "loss": 3.0018, | |
| "step": 149000 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "grad_norm": 0.2459622025489807, | |
| "learning_rate": 0.0006530242409978818, | |
| "loss": 2.9471, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "grad_norm": 0.2830921411514282, | |
| "learning_rate": 0.0006500823723228996, | |
| "loss": 2.9543, | |
| "step": 151000 | |
| }, | |
| { | |
| "epoch": 8.17, | |
| "grad_norm": 0.267229825258255, | |
| "learning_rate": 0.0006471434455165921, | |
| "loss": 2.9631, | |
| "step": 152000 | |
| }, | |
| { | |
| "epoch": 8.23, | |
| "grad_norm": 0.259920597076416, | |
| "learning_rate": 0.0006442015768416098, | |
| "loss": 2.9617, | |
| "step": 153000 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "grad_norm": 0.2159169465303421, | |
| "learning_rate": 0.0006412597081666275, | |
| "loss": 2.9668, | |
| "step": 154000 | |
| }, | |
| { | |
| "epoch": 8.34, | |
| "grad_norm": 0.2374085932970047, | |
| "learning_rate": 0.0006383178394916452, | |
| "loss": 2.9728, | |
| "step": 155000 | |
| }, | |
| { | |
| "epoch": 8.39, | |
| "grad_norm": 0.24743901193141937, | |
| "learning_rate": 0.0006353759708166627, | |
| "loss": 2.9693, | |
| "step": 156000 | |
| }, | |
| { | |
| "epoch": 8.44, | |
| "grad_norm": 0.22295930981636047, | |
| "learning_rate": 0.0006324341021416804, | |
| "loss": 2.9758, | |
| "step": 157000 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "grad_norm": 0.2578998804092407, | |
| "learning_rate": 0.0006294922334666981, | |
| "loss": 2.9743, | |
| "step": 158000 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "grad_norm": 0.23009610176086426, | |
| "learning_rate": 0.0006265533066603907, | |
| "loss": 2.9815, | |
| "step": 159000 | |
| }, | |
| { | |
| "epoch": 8.6, | |
| "grad_norm": 0.2383948415517807, | |
| "learning_rate": 0.0006236143798540833, | |
| "loss": 2.9798, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "grad_norm": 0.2457112967967987, | |
| "learning_rate": 0.000620675453047776, | |
| "loss": 2.9815, | |
| "step": 161000 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "grad_norm": 0.24992822110652924, | |
| "learning_rate": 0.0006177335843727936, | |
| "loss": 2.9797, | |
| "step": 162000 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "grad_norm": 0.24688223004341125, | |
| "learning_rate": 0.0006147946575664862, | |
| "loss": 2.9869, | |
| "step": 163000 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "grad_norm": 0.23152810335159302, | |
| "learning_rate": 0.0006118527888915038, | |
| "loss": 2.9833, | |
| "step": 164000 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "grad_norm": 0.2560078501701355, | |
| "learning_rate": 0.0006089109202165215, | |
| "loss": 2.9869, | |
| "step": 165000 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "grad_norm": 0.2425604909658432, | |
| "learning_rate": 0.0006059690515415393, | |
| "loss": 2.9856, | |
| "step": 166000 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "grad_norm": 0.22484102845191956, | |
| "learning_rate": 0.0006030271828665569, | |
| "loss": 2.9869, | |
| "step": 167000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.40730411942650524, | |
| "eval_loss": 3.3864247798919678, | |
| "eval_runtime": 154.1912, | |
| "eval_samples_per_second": 375.618, | |
| "eval_steps_per_second": 5.869, | |
| "step": 167364 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "grad_norm": 0.24037963151931763, | |
| "learning_rate": 0.0006000882560602495, | |
| "loss": 2.9434, | |
| "step": 168000 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "grad_norm": 0.23906952142715454, | |
| "learning_rate": 0.0005971463873852672, | |
| "loss": 2.9315, | |
| "step": 169000 | |
| }, | |
| { | |
| "epoch": 9.14, | |
| "grad_norm": 0.2632203698158264, | |
| "learning_rate": 0.0005942045187102848, | |
| "loss": 2.9304, | |
| "step": 170000 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "grad_norm": 0.2545742392539978, | |
| "learning_rate": 0.0005912655919039774, | |
| "loss": 2.9354, | |
| "step": 171000 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "grad_norm": 0.2435782551765442, | |
| "learning_rate": 0.0005883237232289951, | |
| "loss": 2.937, | |
| "step": 172000 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "grad_norm": 0.25050055980682373, | |
| "learning_rate": 0.0005853847964226877, | |
| "loss": 2.9435, | |
| "step": 173000 | |
| }, | |
| { | |
| "epoch": 9.36, | |
| "grad_norm": 0.30542534589767456, | |
| "learning_rate": 0.0005824429277477053, | |
| "loss": 2.9454, | |
| "step": 174000 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "grad_norm": 0.23577359318733215, | |
| "learning_rate": 0.000579501059072723, | |
| "loss": 2.9473, | |
| "step": 175000 | |
| }, | |
| { | |
| "epoch": 9.46, | |
| "grad_norm": 0.2357962429523468, | |
| "learning_rate": 0.0005765591903977406, | |
| "loss": 2.9438, | |
| "step": 176000 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "grad_norm": 0.27557680010795593, | |
| "learning_rate": 0.0005736202635914332, | |
| "loss": 2.9519, | |
| "step": 177000 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "grad_norm": 0.2616015672683716, | |
| "learning_rate": 0.0005706813367851259, | |
| "loss": 2.9517, | |
| "step": 178000 | |
| }, | |
| { | |
| "epoch": 9.63, | |
| "grad_norm": 0.2544803321361542, | |
| "learning_rate": 0.0005677394681101436, | |
| "loss": 2.9528, | |
| "step": 179000 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "grad_norm": 0.2523456811904907, | |
| "learning_rate": 0.0005647975994351611, | |
| "loss": 2.9551, | |
| "step": 180000 | |
| }, | |
| { | |
| "epoch": 9.73, | |
| "grad_norm": 0.24971884489059448, | |
| "learning_rate": 0.0005618557307601789, | |
| "loss": 2.9546, | |
| "step": 181000 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "grad_norm": 0.24730221927165985, | |
| "learning_rate": 0.0005589138620851966, | |
| "loss": 2.9569, | |
| "step": 182000 | |
| }, | |
| { | |
| "epoch": 9.84, | |
| "grad_norm": 0.230006605386734, | |
| "learning_rate": 0.0005559749352788892, | |
| "loss": 2.9582, | |
| "step": 183000 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "grad_norm": 0.22560349106788635, | |
| "learning_rate": 0.0005530389503412567, | |
| "loss": 2.9606, | |
| "step": 184000 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "grad_norm": 0.28621554374694824, | |
| "learning_rate": 0.0005500970816662745, | |
| "loss": 2.9651, | |
| "step": 185000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.4079115901065721, | |
| "eval_loss": 3.361621856689453, | |
| "eval_runtime": 154.014, | |
| "eval_samples_per_second": 376.05, | |
| "eval_steps_per_second": 5.876, | |
| "step": 185960 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.2537492513656616, | |
| "learning_rate": 0.0005471552129912922, | |
| "loss": 2.9625, | |
| "step": 186000 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "grad_norm": 0.23095977306365967, | |
| "learning_rate": 0.0005442133443163097, | |
| "loss": 2.901, | |
| "step": 187000 | |
| }, | |
| { | |
| "epoch": 10.11, | |
| "grad_norm": 0.2696865200996399, | |
| "learning_rate": 0.0005412714756413274, | |
| "loss": 2.9027, | |
| "step": 188000 | |
| }, | |
| { | |
| "epoch": 10.16, | |
| "grad_norm": 0.26662537455558777, | |
| "learning_rate": 0.0005383325488350201, | |
| "loss": 2.912, | |
| "step": 189000 | |
| }, | |
| { | |
| "epoch": 10.22, | |
| "grad_norm": 0.24653853476047516, | |
| "learning_rate": 0.0005353906801600376, | |
| "loss": 2.9081, | |
| "step": 190000 | |
| }, | |
| { | |
| "epoch": 10.27, | |
| "grad_norm": 0.2434544712305069, | |
| "learning_rate": 0.0005324488114850553, | |
| "loss": 2.9184, | |
| "step": 191000 | |
| }, | |
| { | |
| "epoch": 10.32, | |
| "grad_norm": 0.25054100155830383, | |
| "learning_rate": 0.000529506942810073, | |
| "loss": 2.9183, | |
| "step": 192000 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "grad_norm": 0.27828437089920044, | |
| "learning_rate": 0.0005265650741350905, | |
| "loss": 2.9214, | |
| "step": 193000 | |
| }, | |
| { | |
| "epoch": 10.43, | |
| "grad_norm": 0.2823689579963684, | |
| "learning_rate": 0.0005236290891974582, | |
| "loss": 2.9229, | |
| "step": 194000 | |
| }, | |
| { | |
| "epoch": 10.49, | |
| "grad_norm": 0.26634788513183594, | |
| "learning_rate": 0.0005206872205224759, | |
| "loss": 2.9256, | |
| "step": 195000 | |
| }, | |
| { | |
| "epoch": 10.54, | |
| "grad_norm": 0.314277708530426, | |
| "learning_rate": 0.0005177453518474935, | |
| "loss": 2.9326, | |
| "step": 196000 | |
| }, | |
| { | |
| "epoch": 10.59, | |
| "grad_norm": 0.24600285291671753, | |
| "learning_rate": 0.0005148034831725112, | |
| "loss": 2.9236, | |
| "step": 197000 | |
| }, | |
| { | |
| "epoch": 10.65, | |
| "grad_norm": 0.2303924709558487, | |
| "learning_rate": 0.0005118645563662038, | |
| "loss": 2.9316, | |
| "step": 198000 | |
| }, | |
| { | |
| "epoch": 10.7, | |
| "grad_norm": 0.23870837688446045, | |
| "learning_rate": 0.0005089226876912215, | |
| "loss": 2.9352, | |
| "step": 199000 | |
| }, | |
| { | |
| "epoch": 10.76, | |
| "grad_norm": 0.2509617805480957, | |
| "learning_rate": 0.0005059808190162391, | |
| "loss": 2.9302, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 10.81, | |
| "grad_norm": 0.28766971826553345, | |
| "learning_rate": 0.0005030389503412568, | |
| "loss": 2.9309, | |
| "step": 201000 | |
| }, | |
| { | |
| "epoch": 10.86, | |
| "grad_norm": 0.26924803853034973, | |
| "learning_rate": 0.0005001000235349494, | |
| "loss": 2.9381, | |
| "step": 202000 | |
| }, | |
| { | |
| "epoch": 10.92, | |
| "grad_norm": 0.26498013734817505, | |
| "learning_rate": 0.0004971581548599671, | |
| "loss": 2.9364, | |
| "step": 203000 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "grad_norm": 0.2591700255870819, | |
| "learning_rate": 0.0004942162861849847, | |
| "loss": 2.9407, | |
| "step": 204000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.40898988079613996, | |
| "eval_loss": 3.3727080821990967, | |
| "eval_runtime": 154.0532, | |
| "eval_samples_per_second": 375.955, | |
| "eval_steps_per_second": 5.875, | |
| "step": 204556 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "grad_norm": 0.2849581837654114, | |
| "learning_rate": 0.0004912744175100024, | |
| "loss": 2.9093, | |
| "step": 205000 | |
| }, | |
| { | |
| "epoch": 11.08, | |
| "grad_norm": 0.26025569438934326, | |
| "learning_rate": 0.00048833843257237, | |
| "loss": 2.8745, | |
| "step": 206000 | |
| }, | |
| { | |
| "epoch": 11.13, | |
| "grad_norm": 0.2565370202064514, | |
| "learning_rate": 0.00048539656389738763, | |
| "loss": 2.8846, | |
| "step": 207000 | |
| }, | |
| { | |
| "epoch": 11.19, | |
| "grad_norm": 0.29611414670944214, | |
| "learning_rate": 0.00048245763709108024, | |
| "loss": 2.8862, | |
| "step": 208000 | |
| }, | |
| { | |
| "epoch": 11.24, | |
| "grad_norm": 0.23962269723415375, | |
| "learning_rate": 0.0004795157684160979, | |
| "loss": 2.8923, | |
| "step": 209000 | |
| }, | |
| { | |
| "epoch": 11.29, | |
| "grad_norm": 0.23374328017234802, | |
| "learning_rate": 0.0004765768416097905, | |
| "loss": 2.895, | |
| "step": 210000 | |
| }, | |
| { | |
| "epoch": 11.35, | |
| "grad_norm": 0.2842772305011749, | |
| "learning_rate": 0.0004736349729348082, | |
| "loss": 2.8978, | |
| "step": 211000 | |
| }, | |
| { | |
| "epoch": 11.4, | |
| "grad_norm": 0.2555365562438965, | |
| "learning_rate": 0.0004706960461285008, | |
| "loss": 2.8982, | |
| "step": 212000 | |
| }, | |
| { | |
| "epoch": 11.45, | |
| "grad_norm": 0.26069942116737366, | |
| "learning_rate": 0.0004677541774535185, | |
| "loss": 2.9035, | |
| "step": 213000 | |
| }, | |
| { | |
| "epoch": 11.51, | |
| "grad_norm": 0.2919773459434509, | |
| "learning_rate": 0.00046481230877853615, | |
| "loss": 2.9057, | |
| "step": 214000 | |
| }, | |
| { | |
| "epoch": 11.56, | |
| "grad_norm": 0.26192766427993774, | |
| "learning_rate": 0.0004618704401035538, | |
| "loss": 2.9039, | |
| "step": 215000 | |
| }, | |
| { | |
| "epoch": 11.62, | |
| "grad_norm": 0.2489086240530014, | |
| "learning_rate": 0.0004589285714285714, | |
| "loss": 2.9093, | |
| "step": 216000 | |
| }, | |
| { | |
| "epoch": 11.67, | |
| "grad_norm": 0.26663750410079956, | |
| "learning_rate": 0.00045598964462226406, | |
| "loss": 2.9134, | |
| "step": 217000 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "grad_norm": 0.26307132840156555, | |
| "learning_rate": 0.00045304777594728173, | |
| "loss": 2.9088, | |
| "step": 218000 | |
| }, | |
| { | |
| "epoch": 11.78, | |
| "grad_norm": 0.2765272259712219, | |
| "learning_rate": 0.00045010884914097434, | |
| "loss": 2.9124, | |
| "step": 219000 | |
| }, | |
| { | |
| "epoch": 11.83, | |
| "grad_norm": 0.2534734010696411, | |
| "learning_rate": 0.000447166980465992, | |
| "loss": 2.9139, | |
| "step": 220000 | |
| }, | |
| { | |
| "epoch": 11.88, | |
| "grad_norm": 0.2891595661640167, | |
| "learning_rate": 0.0004442251117910097, | |
| "loss": 2.9135, | |
| "step": 221000 | |
| }, | |
| { | |
| "epoch": 11.94, | |
| "grad_norm": 0.27639132738113403, | |
| "learning_rate": 0.0004412861849847023, | |
| "loss": 2.9144, | |
| "step": 222000 | |
| }, | |
| { | |
| "epoch": 11.99, | |
| "grad_norm": 0.2949223518371582, | |
| "learning_rate": 0.0004383443163097199, | |
| "loss": 2.9186, | |
| "step": 223000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.4090768830667313, | |
| "eval_loss": 3.386943817138672, | |
| "eval_runtime": 154.0531, | |
| "eval_samples_per_second": 375.955, | |
| "eval_steps_per_second": 5.875, | |
| "step": 223152 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "grad_norm": 0.2919670641422272, | |
| "learning_rate": 0.0004354053895034126, | |
| "loss": 2.8624, | |
| "step": 224000 | |
| }, | |
| { | |
| "epoch": 12.1, | |
| "grad_norm": 0.2778524160385132, | |
| "learning_rate": 0.0004324664626971052, | |
| "loss": 2.8607, | |
| "step": 225000 | |
| }, | |
| { | |
| "epoch": 12.15, | |
| "grad_norm": 0.2688823640346527, | |
| "learning_rate": 0.00042952459402212286, | |
| "loss": 2.8637, | |
| "step": 226000 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "grad_norm": 0.26661086082458496, | |
| "learning_rate": 0.0004265827253471405, | |
| "loss": 2.8716, | |
| "step": 227000 | |
| }, | |
| { | |
| "epoch": 12.26, | |
| "grad_norm": 0.2500409483909607, | |
| "learning_rate": 0.00042364085667215815, | |
| "loss": 2.8719, | |
| "step": 228000 | |
| }, | |
| { | |
| "epoch": 12.31, | |
| "grad_norm": 0.24592752754688263, | |
| "learning_rate": 0.0004207019298658508, | |
| "loss": 2.8779, | |
| "step": 229000 | |
| }, | |
| { | |
| "epoch": 12.37, | |
| "grad_norm": 0.2966313064098358, | |
| "learning_rate": 0.00041776006119086844, | |
| "loss": 2.8773, | |
| "step": 230000 | |
| }, | |
| { | |
| "epoch": 12.42, | |
| "grad_norm": 0.2876270115375519, | |
| "learning_rate": 0.0004148181925158861, | |
| "loss": 2.8777, | |
| "step": 231000 | |
| }, | |
| { | |
| "epoch": 12.48, | |
| "grad_norm": 0.25883278250694275, | |
| "learning_rate": 0.0004118792657095787, | |
| "loss": 2.8838, | |
| "step": 232000 | |
| }, | |
| { | |
| "epoch": 12.53, | |
| "grad_norm": 0.25680166482925415, | |
| "learning_rate": 0.0004089373970345964, | |
| "loss": 2.8847, | |
| "step": 233000 | |
| }, | |
| { | |
| "epoch": 12.58, | |
| "grad_norm": 0.28966328501701355, | |
| "learning_rate": 0.000405998470228289, | |
| "loss": 2.8847, | |
| "step": 234000 | |
| }, | |
| { | |
| "epoch": 12.64, | |
| "grad_norm": 0.2587486803531647, | |
| "learning_rate": 0.0004030566015533066, | |
| "loss": 2.8868, | |
| "step": 235000 | |
| }, | |
| { | |
| "epoch": 12.69, | |
| "grad_norm": 0.2806561291217804, | |
| "learning_rate": 0.00040011473287832435, | |
| "loss": 2.8887, | |
| "step": 236000 | |
| }, | |
| { | |
| "epoch": 12.74, | |
| "grad_norm": 0.25261130928993225, | |
| "learning_rate": 0.00039717580607201696, | |
| "loss": 2.8888, | |
| "step": 237000 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "grad_norm": 0.2473638653755188, | |
| "learning_rate": 0.00039423393739703463, | |
| "loss": 2.8931, | |
| "step": 238000 | |
| }, | |
| { | |
| "epoch": 12.85, | |
| "grad_norm": 0.2695077061653137, | |
| "learning_rate": 0.00039129206872205225, | |
| "loss": 2.8937, | |
| "step": 239000 | |
| }, | |
| { | |
| "epoch": 12.91, | |
| "grad_norm": 0.26754283905029297, | |
| "learning_rate": 0.00038835020004706987, | |
| "loss": 2.8929, | |
| "step": 240000 | |
| }, | |
| { | |
| "epoch": 12.96, | |
| "grad_norm": 0.2615933120250702, | |
| "learning_rate": 0.00038541127324076254, | |
| "loss": 2.8974, | |
| "step": 241000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.4096338991481772, | |
| "eval_loss": 3.3688578605651855, | |
| "eval_runtime": 154.4918, | |
| "eval_samples_per_second": 374.887, | |
| "eval_steps_per_second": 5.858, | |
| "step": 241748 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "grad_norm": 0.2606102526187897, | |
| "learning_rate": 0.00038246940456578016, | |
| "loss": 2.8775, | |
| "step": 242000 | |
| }, | |
| { | |
| "epoch": 13.07, | |
| "grad_norm": 0.27312567830085754, | |
| "learning_rate": 0.0003795275358907979, | |
| "loss": 2.837, | |
| "step": 243000 | |
| }, | |
| { | |
| "epoch": 13.12, | |
| "grad_norm": 0.2866615056991577, | |
| "learning_rate": 0.0003765886090844905, | |
| "loss": 2.8441, | |
| "step": 244000 | |
| }, | |
| { | |
| "epoch": 13.17, | |
| "grad_norm": 0.26024916768074036, | |
| "learning_rate": 0.0003736467404095081, | |
| "loss": 2.8453, | |
| "step": 245000 | |
| }, | |
| { | |
| "epoch": 13.23, | |
| "grad_norm": 0.3037625551223755, | |
| "learning_rate": 0.0003707107554718757, | |
| "loss": 2.849, | |
| "step": 246000 | |
| }, | |
| { | |
| "epoch": 13.28, | |
| "grad_norm": 0.25507059693336487, | |
| "learning_rate": 0.0003677688867968934, | |
| "loss": 2.8511, | |
| "step": 247000 | |
| }, | |
| { | |
| "epoch": 13.34, | |
| "grad_norm": 0.26291272044181824, | |
| "learning_rate": 0.00036482701812191106, | |
| "loss": 2.8608, | |
| "step": 248000 | |
| }, | |
| { | |
| "epoch": 13.39, | |
| "grad_norm": 0.2695772349834442, | |
| "learning_rate": 0.0003618851494469287, | |
| "loss": 2.8596, | |
| "step": 249000 | |
| }, | |
| { | |
| "epoch": 13.44, | |
| "grad_norm": 0.2734662592411041, | |
| "learning_rate": 0.00035894622264062134, | |
| "loss": 2.8626, | |
| "step": 250000 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "grad_norm": 0.34274160861968994, | |
| "learning_rate": 0.00035600435396563896, | |
| "loss": 2.8624, | |
| "step": 251000 | |
| }, | |
| { | |
| "epoch": 13.55, | |
| "grad_norm": 0.2600579857826233, | |
| "learning_rate": 0.00035306248529065663, | |
| "loss": 2.8641, | |
| "step": 252000 | |
| }, | |
| { | |
| "epoch": 13.61, | |
| "grad_norm": 0.27528443932533264, | |
| "learning_rate": 0.0003501235584843493, | |
| "loss": 2.8663, | |
| "step": 253000 | |
| }, | |
| { | |
| "epoch": 13.66, | |
| "grad_norm": 0.26855769753456116, | |
| "learning_rate": 0.0003471816898093669, | |
| "loss": 2.872, | |
| "step": 254000 | |
| }, | |
| { | |
| "epoch": 13.71, | |
| "grad_norm": 0.26368248462677, | |
| "learning_rate": 0.0003442427630030596, | |
| "loss": 2.8674, | |
| "step": 255000 | |
| }, | |
| { | |
| "epoch": 13.77, | |
| "grad_norm": 0.27387478947639465, | |
| "learning_rate": 0.0003413008943280772, | |
| "loss": 2.8665, | |
| "step": 256000 | |
| }, | |
| { | |
| "epoch": 13.82, | |
| "grad_norm": 0.30804863572120667, | |
| "learning_rate": 0.0003383590256530948, | |
| "loss": 2.8721, | |
| "step": 257000 | |
| }, | |
| { | |
| "epoch": 13.87, | |
| "grad_norm": 0.3087449371814728, | |
| "learning_rate": 0.0003354200988467875, | |
| "loss": 2.8765, | |
| "step": 258000 | |
| }, | |
| { | |
| "epoch": 13.93, | |
| "grad_norm": 0.26736271381378174, | |
| "learning_rate": 0.0003324782301718051, | |
| "loss": 2.8768, | |
| "step": 259000 | |
| }, | |
| { | |
| "epoch": 13.98, | |
| "grad_norm": 0.3191315233707428, | |
| "learning_rate": 0.00032953636149682283, | |
| "loss": 2.8733, | |
| "step": 260000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.40978962985337464, | |
| "eval_loss": 3.381140947341919, | |
| "eval_runtime": 153.947, | |
| "eval_samples_per_second": 376.214, | |
| "eval_steps_per_second": 5.879, | |
| "step": 260344 | |
| }, | |
| { | |
| "epoch": 14.04, | |
| "grad_norm": 0.2935682237148285, | |
| "learning_rate": 0.00032659743469051544, | |
| "loss": 2.8349, | |
| "step": 261000 | |
| }, | |
| { | |
| "epoch": 14.09, | |
| "grad_norm": 0.32009586691856384, | |
| "learning_rate": 0.0003236555660155331, | |
| "loss": 2.8245, | |
| "step": 262000 | |
| }, | |
| { | |
| "epoch": 14.14, | |
| "grad_norm": 0.2821045219898224, | |
| "learning_rate": 0.00032071369734055073, | |
| "loss": 2.8243, | |
| "step": 263000 | |
| }, | |
| { | |
| "epoch": 14.2, | |
| "grad_norm": 0.2877224087715149, | |
| "learning_rate": 0.00031777477053424334, | |
| "loss": 2.8294, | |
| "step": 264000 | |
| }, | |
| { | |
| "epoch": 14.25, | |
| "grad_norm": 0.28222963213920593, | |
| "learning_rate": 0.000314835843727936, | |
| "loss": 2.8345, | |
| "step": 265000 | |
| }, | |
| { | |
| "epoch": 14.3, | |
| "grad_norm": 0.2666056156158447, | |
| "learning_rate": 0.0003118939750529536, | |
| "loss": 2.8345, | |
| "step": 266000 | |
| }, | |
| { | |
| "epoch": 14.36, | |
| "grad_norm": 0.26124948263168335, | |
| "learning_rate": 0.0003089521063779713, | |
| "loss": 2.8419, | |
| "step": 267000 | |
| }, | |
| { | |
| "epoch": 14.41, | |
| "grad_norm": 0.3096939027309418, | |
| "learning_rate": 0.00030601023770298897, | |
| "loss": 2.8415, | |
| "step": 268000 | |
| }, | |
| { | |
| "epoch": 14.47, | |
| "grad_norm": 0.27581918239593506, | |
| "learning_rate": 0.0003030683690280066, | |
| "loss": 2.8435, | |
| "step": 269000 | |
| }, | |
| { | |
| "epoch": 14.52, | |
| "grad_norm": 0.29562026262283325, | |
| "learning_rate": 0.00030012944222169925, | |
| "loss": 2.8451, | |
| "step": 270000 | |
| }, | |
| { | |
| "epoch": 14.57, | |
| "grad_norm": 0.30838248133659363, | |
| "learning_rate": 0.0002971875735467169, | |
| "loss": 2.8491, | |
| "step": 271000 | |
| }, | |
| { | |
| "epoch": 14.63, | |
| "grad_norm": 0.2891804873943329, | |
| "learning_rate": 0.00029424864674040954, | |
| "loss": 2.8485, | |
| "step": 272000 | |
| }, | |
| { | |
| "epoch": 14.68, | |
| "grad_norm": 0.2642837166786194, | |
| "learning_rate": 0.00029130677806542716, | |
| "loss": 2.8478, | |
| "step": 273000 | |
| }, | |
| { | |
| "epoch": 14.73, | |
| "grad_norm": 0.3049638271331787, | |
| "learning_rate": 0.0002883649093904448, | |
| "loss": 2.8519, | |
| "step": 274000 | |
| }, | |
| { | |
| "epoch": 14.79, | |
| "grad_norm": 0.26046690344810486, | |
| "learning_rate": 0.0002854230407154625, | |
| "loss": 2.8514, | |
| "step": 275000 | |
| }, | |
| { | |
| "epoch": 14.84, | |
| "grad_norm": 0.2902365028858185, | |
| "learning_rate": 0.0002824841139091551, | |
| "loss": 2.8505, | |
| "step": 276000 | |
| }, | |
| { | |
| "epoch": 14.9, | |
| "grad_norm": 0.2648141384124756, | |
| "learning_rate": 0.0002795422452341728, | |
| "loss": 2.8525, | |
| "step": 277000 | |
| }, | |
| { | |
| "epoch": 14.95, | |
| "grad_norm": 0.27711859345436096, | |
| "learning_rate": 0.0002766003765591904, | |
| "loss": 2.8547, | |
| "step": 278000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.41021328724206096, | |
| "eval_loss": 3.3861238956451416, | |
| "eval_runtime": 154.0149, | |
| "eval_samples_per_second": 376.048, | |
| "eval_steps_per_second": 5.876, | |
| "step": 278940 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.27470868825912476, | |
| "learning_rate": 0.000273658507884208, | |
| "loss": 2.8524, | |
| "step": 279000 | |
| }, | |
| { | |
| "epoch": 15.06, | |
| "grad_norm": 0.29804325103759766, | |
| "learning_rate": 0.0002707225229465757, | |
| "loss": 2.8018, | |
| "step": 280000 | |
| }, | |
| { | |
| "epoch": 15.11, | |
| "grad_norm": 0.3048715591430664, | |
| "learning_rate": 0.0002677806542715933, | |
| "loss": 2.8103, | |
| "step": 281000 | |
| }, | |
| { | |
| "epoch": 15.16, | |
| "grad_norm": 0.2895219624042511, | |
| "learning_rate": 0.00026483878559661097, | |
| "loss": 2.812, | |
| "step": 282000 | |
| }, | |
| { | |
| "epoch": 15.22, | |
| "grad_norm": 0.31599316000938416, | |
| "learning_rate": 0.00026189691692162865, | |
| "loss": 2.82, | |
| "step": 283000 | |
| }, | |
| { | |
| "epoch": 15.27, | |
| "grad_norm": 0.2961556315422058, | |
| "learning_rate": 0.00025895504824664626, | |
| "loss": 2.8214, | |
| "step": 284000 | |
| }, | |
| { | |
| "epoch": 15.33, | |
| "grad_norm": 0.3021540343761444, | |
| "learning_rate": 0.00025601612144033893, | |
| "loss": 2.8174, | |
| "step": 285000 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "grad_norm": 0.2883637845516205, | |
| "learning_rate": 0.00025307425276535655, | |
| "loss": 2.8197, | |
| "step": 286000 | |
| }, | |
| { | |
| "epoch": 15.43, | |
| "grad_norm": 0.2952839136123657, | |
| "learning_rate": 0.0002501323840903742, | |
| "loss": 2.8187, | |
| "step": 287000 | |
| }, | |
| { | |
| "epoch": 15.49, | |
| "grad_norm": 0.2684228718280792, | |
| "learning_rate": 0.00024719051541539184, | |
| "loss": 2.826, | |
| "step": 288000 | |
| }, | |
| { | |
| "epoch": 15.54, | |
| "grad_norm": 0.31188449263572693, | |
| "learning_rate": 0.0002442515886090845, | |
| "loss": 2.8287, | |
| "step": 289000 | |
| }, | |
| { | |
| "epoch": 15.59, | |
| "grad_norm": 0.2904526889324188, | |
| "learning_rate": 0.00024130971993410215, | |
| "loss": 2.8232, | |
| "step": 290000 | |
| }, | |
| { | |
| "epoch": 15.65, | |
| "grad_norm": 0.31149864196777344, | |
| "learning_rate": 0.0002383678512591198, | |
| "loss": 2.8305, | |
| "step": 291000 | |
| }, | |
| { | |
| "epoch": 15.7, | |
| "grad_norm": 0.30907660722732544, | |
| "learning_rate": 0.00023542892445281243, | |
| "loss": 2.8319, | |
| "step": 292000 | |
| }, | |
| { | |
| "epoch": 15.76, | |
| "grad_norm": 0.2970659136772156, | |
| "learning_rate": 0.00023248705577783008, | |
| "loss": 2.8319, | |
| "step": 293000 | |
| }, | |
| { | |
| "epoch": 15.81, | |
| "grad_norm": 0.32185524702072144, | |
| "learning_rate": 0.00022954812897152272, | |
| "loss": 2.8334, | |
| "step": 294000 | |
| }, | |
| { | |
| "epoch": 15.86, | |
| "grad_norm": 0.2747587561607361, | |
| "learning_rate": 0.00022660626029654036, | |
| "loss": 2.8351, | |
| "step": 295000 | |
| }, | |
| { | |
| "epoch": 15.92, | |
| "grad_norm": 0.32215383648872375, | |
| "learning_rate": 0.000223664391621558, | |
| "loss": 2.8343, | |
| "step": 296000 | |
| }, | |
| { | |
| "epoch": 15.97, | |
| "grad_norm": 0.30099472403526306, | |
| "learning_rate": 0.00022072252294657566, | |
| "loss": 2.8352, | |
| "step": 297000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.4099246009434271, | |
| "eval_loss": 3.3998661041259766, | |
| "eval_runtime": 154.387, | |
| "eval_samples_per_second": 375.142, | |
| "eval_steps_per_second": 5.862, | |
| "step": 297536 | |
| }, | |
| { | |
| "epoch": 16.02, | |
| "grad_norm": 0.3042096495628357, | |
| "learning_rate": 0.0002177835961402683, | |
| "loss": 2.8115, | |
| "step": 298000 | |
| }, | |
| { | |
| "epoch": 16.08, | |
| "grad_norm": 0.29497480392456055, | |
| "learning_rate": 0.00021484172746528597, | |
| "loss": 2.7915, | |
| "step": 299000 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "grad_norm": 0.2896369993686676, | |
| "learning_rate": 0.0002118998587903036, | |
| "loss": 2.7949, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 16.19, | |
| "grad_norm": 0.3014323115348816, | |
| "learning_rate": 0.00020896093198399625, | |
| "loss": 2.7968, | |
| "step": 301000 | |
| }, | |
| { | |
| "epoch": 16.24, | |
| "grad_norm": 0.2915323078632355, | |
| "learning_rate": 0.00020601906330901387, | |
| "loss": 2.7961, | |
| "step": 302000 | |
| }, | |
| { | |
| "epoch": 16.29, | |
| "grad_norm": 0.28494659066200256, | |
| "learning_rate": 0.00020307719463403154, | |
| "loss": 2.7993, | |
| "step": 303000 | |
| }, | |
| { | |
| "epoch": 16.35, | |
| "grad_norm": 0.2689177095890045, | |
| "learning_rate": 0.00020013826782772418, | |
| "loss": 2.8054, | |
| "step": 304000 | |
| }, | |
| { | |
| "epoch": 16.4, | |
| "grad_norm": 0.290145605802536, | |
| "learning_rate": 0.00019719639915274182, | |
| "loss": 2.808, | |
| "step": 305000 | |
| }, | |
| { | |
| "epoch": 16.46, | |
| "grad_norm": 0.306331992149353, | |
| "learning_rate": 0.00019425747234643446, | |
| "loss": 2.8077, | |
| "step": 306000 | |
| }, | |
| { | |
| "epoch": 16.51, | |
| "grad_norm": 0.28849080204963684, | |
| "learning_rate": 0.0001913156036714521, | |
| "loss": 2.808, | |
| "step": 307000 | |
| }, | |
| { | |
| "epoch": 16.56, | |
| "grad_norm": 0.33114251494407654, | |
| "learning_rate": 0.00018837373499646975, | |
| "loss": 2.8096, | |
| "step": 308000 | |
| }, | |
| { | |
| "epoch": 16.62, | |
| "grad_norm": 0.32337045669555664, | |
| "learning_rate": 0.0001854348081901624, | |
| "loss": 2.8101, | |
| "step": 309000 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "grad_norm": 0.3087981939315796, | |
| "learning_rate": 0.00018249293951518004, | |
| "loss": 2.8112, | |
| "step": 310000 | |
| }, | |
| { | |
| "epoch": 16.72, | |
| "grad_norm": 0.2813110947608948, | |
| "learning_rate": 0.00017955401270887267, | |
| "loss": 2.8153, | |
| "step": 311000 | |
| }, | |
| { | |
| "epoch": 16.78, | |
| "grad_norm": 0.28998225927352905, | |
| "learning_rate": 0.00017661214403389035, | |
| "loss": 2.815, | |
| "step": 312000 | |
| }, | |
| { | |
| "epoch": 16.83, | |
| "grad_norm": 0.29306545853614807, | |
| "learning_rate": 0.00017367321722758298, | |
| "loss": 2.8161, | |
| "step": 313000 | |
| }, | |
| { | |
| "epoch": 16.89, | |
| "grad_norm": 0.3053112328052521, | |
| "learning_rate": 0.0001707313485526006, | |
| "loss": 2.8119, | |
| "step": 314000 | |
| }, | |
| { | |
| "epoch": 16.94, | |
| "grad_norm": 0.3361365795135498, | |
| "learning_rate": 0.00016778947987761828, | |
| "loss": 2.8147, | |
| "step": 315000 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "grad_norm": 0.30090630054473877, | |
| "learning_rate": 0.00016485349493998588, | |
| "loss": 2.8173, | |
| "step": 316000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.4103711006270949, | |
| "eval_loss": 3.3992600440979004, | |
| "eval_runtime": 154.0327, | |
| "eval_samples_per_second": 376.005, | |
| "eval_steps_per_second": 5.875, | |
| "step": 316132 | |
| }, | |
| { | |
| "epoch": 17.05, | |
| "grad_norm": 0.28460776805877686, | |
| "learning_rate": 0.00016191162626500352, | |
| "loss": 2.7809, | |
| "step": 317000 | |
| }, | |
| { | |
| "epoch": 17.1, | |
| "grad_norm": 0.2945604920387268, | |
| "learning_rate": 0.0001589697575900212, | |
| "loss": 2.7761, | |
| "step": 318000 | |
| }, | |
| { | |
| "epoch": 17.15, | |
| "grad_norm": 0.2811717987060547, | |
| "learning_rate": 0.00015603083078371383, | |
| "loss": 2.7791, | |
| "step": 319000 | |
| }, | |
| { | |
| "epoch": 17.21, | |
| "grad_norm": 0.28703057765960693, | |
| "learning_rate": 0.00015308896210873148, | |
| "loss": 2.785, | |
| "step": 320000 | |
| }, | |
| { | |
| "epoch": 17.26, | |
| "grad_norm": 0.2974686622619629, | |
| "learning_rate": 0.00015014709343374912, | |
| "loss": 2.7862, | |
| "step": 321000 | |
| }, | |
| { | |
| "epoch": 17.32, | |
| "grad_norm": 0.29753968119621277, | |
| "learning_rate": 0.00014720522475876677, | |
| "loss": 2.7875, | |
| "step": 322000 | |
| }, | |
| { | |
| "epoch": 17.37, | |
| "grad_norm": 0.29779624938964844, | |
| "learning_rate": 0.0001442662979524594, | |
| "loss": 2.7876, | |
| "step": 323000 | |
| }, | |
| { | |
| "epoch": 17.42, | |
| "grad_norm": 0.281829833984375, | |
| "learning_rate": 0.00014132442927747705, | |
| "loss": 2.7909, | |
| "step": 324000 | |
| }, | |
| { | |
| "epoch": 17.48, | |
| "grad_norm": 0.3030970096588135, | |
| "learning_rate": 0.0001383825606024947, | |
| "loss": 2.7883, | |
| "step": 325000 | |
| }, | |
| { | |
| "epoch": 17.53, | |
| "grad_norm": 0.28257811069488525, | |
| "learning_rate": 0.00013544363379618734, | |
| "loss": 2.793, | |
| "step": 326000 | |
| }, | |
| { | |
| "epoch": 17.58, | |
| "grad_norm": 0.3062205910682678, | |
| "learning_rate": 0.00013250176512120498, | |
| "loss": 2.7923, | |
| "step": 327000 | |
| }, | |
| { | |
| "epoch": 17.64, | |
| "grad_norm": 0.29610031843185425, | |
| "learning_rate": 0.00012956283831489762, | |
| "loss": 2.7947, | |
| "step": 328000 | |
| }, | |
| { | |
| "epoch": 17.69, | |
| "grad_norm": 0.31379351019859314, | |
| "learning_rate": 0.0001266209696399153, | |
| "loss": 2.793, | |
| "step": 329000 | |
| }, | |
| { | |
| "epoch": 17.75, | |
| "grad_norm": 0.29375967383384705, | |
| "learning_rate": 0.00012367910096493294, | |
| "loss": 2.7971, | |
| "step": 330000 | |
| }, | |
| { | |
| "epoch": 17.8, | |
| "grad_norm": 0.30711761116981506, | |
| "learning_rate": 0.00012073723228995057, | |
| "loss": 2.7963, | |
| "step": 331000 | |
| }, | |
| { | |
| "epoch": 17.85, | |
| "grad_norm": 0.29258862137794495, | |
| "learning_rate": 0.00011780124735231819, | |
| "loss": 2.7937, | |
| "step": 332000 | |
| }, | |
| { | |
| "epoch": 17.91, | |
| "grad_norm": 0.2702758014202118, | |
| "learning_rate": 0.00011485937867733585, | |
| "loss": 2.7942, | |
| "step": 333000 | |
| }, | |
| { | |
| "epoch": 17.96, | |
| "grad_norm": 0.2748437523841858, | |
| "learning_rate": 0.00011191751000235349, | |
| "loss": 2.8005, | |
| "step": 334000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.4099377016714312, | |
| "eval_loss": 3.4154224395751953, | |
| "eval_runtime": 154.4547, | |
| "eval_samples_per_second": 374.977, | |
| "eval_steps_per_second": 5.859, | |
| "step": 334728 | |
| }, | |
| { | |
| "epoch": 18.01, | |
| "grad_norm": 0.29431185126304626, | |
| "learning_rate": 0.00010897858319604613, | |
| "loss": 2.7895, | |
| "step": 335000 | |
| }, | |
| { | |
| "epoch": 18.07, | |
| "grad_norm": 0.3129688799381256, | |
| "learning_rate": 0.00010603965638973876, | |
| "loss": 2.7669, | |
| "step": 336000 | |
| }, | |
| { | |
| "epoch": 18.12, | |
| "grad_norm": 0.29429852962493896, | |
| "learning_rate": 0.00010309778771475643, | |
| "loss": 2.7668, | |
| "step": 337000 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "grad_norm": 0.30273085832595825, | |
| "learning_rate": 0.00010015591903977407, | |
| "loss": 2.7695, | |
| "step": 338000 | |
| }, | |
| { | |
| "epoch": 18.23, | |
| "grad_norm": 0.3032689392566681, | |
| "learning_rate": 9.72169922334667e-05, | |
| "loss": 2.7712, | |
| "step": 339000 | |
| }, | |
| { | |
| "epoch": 18.28, | |
| "grad_norm": 0.3074301779270172, | |
| "learning_rate": 9.427512355848434e-05, | |
| "loss": 2.7691, | |
| "step": 340000 | |
| }, | |
| { | |
| "epoch": 18.34, | |
| "grad_norm": 0.30292806029319763, | |
| "learning_rate": 9.1333254883502e-05, | |
| "loss": 2.7725, | |
| "step": 341000 | |
| }, | |
| { | |
| "epoch": 18.39, | |
| "grad_norm": 0.31416478753089905, | |
| "learning_rate": 8.839432807719464e-05, | |
| "loss": 2.7718, | |
| "step": 342000 | |
| }, | |
| { | |
| "epoch": 18.44, | |
| "grad_norm": 0.3108297884464264, | |
| "learning_rate": 8.545245940221228e-05, | |
| "loss": 2.7724, | |
| "step": 343000 | |
| }, | |
| { | |
| "epoch": 18.5, | |
| "grad_norm": 0.31622520089149475, | |
| "learning_rate": 8.251059072722994e-05, | |
| "loss": 2.7771, | |
| "step": 344000 | |
| }, | |
| { | |
| "epoch": 18.55, | |
| "grad_norm": 0.2999779284000397, | |
| "learning_rate": 7.956872205224759e-05, | |
| "loss": 2.7739, | |
| "step": 345000 | |
| }, | |
| { | |
| "epoch": 18.61, | |
| "grad_norm": 0.28529179096221924, | |
| "learning_rate": 7.662685337726524e-05, | |
| "loss": 2.7776, | |
| "step": 346000 | |
| }, | |
| { | |
| "epoch": 18.66, | |
| "grad_norm": 0.3071877956390381, | |
| "learning_rate": 7.368792657095789e-05, | |
| "loss": 2.7764, | |
| "step": 347000 | |
| }, | |
| { | |
| "epoch": 18.71, | |
| "grad_norm": 0.3046530783176422, | |
| "learning_rate": 7.074605789597552e-05, | |
| "loss": 2.776, | |
| "step": 348000 | |
| }, | |
| { | |
| "epoch": 18.77, | |
| "grad_norm": 0.3025725781917572, | |
| "learning_rate": 6.780418922099317e-05, | |
| "loss": 2.7787, | |
| "step": 349000 | |
| }, | |
| { | |
| "epoch": 18.82, | |
| "grad_norm": 0.3011409044265747, | |
| "learning_rate": 6.48652624146858e-05, | |
| "loss": 2.7794, | |
| "step": 350000 | |
| }, | |
| { | |
| "epoch": 18.88, | |
| "grad_norm": 0.2958693206310272, | |
| "learning_rate": 6.192339373970346e-05, | |
| "loss": 2.7778, | |
| "step": 351000 | |
| }, | |
| { | |
| "epoch": 18.93, | |
| "grad_norm": 0.3003888726234436, | |
| "learning_rate": 5.898152506472111e-05, | |
| "loss": 2.7786, | |
| "step": 352000 | |
| }, | |
| { | |
| "epoch": 18.98, | |
| "grad_norm": 0.29960742592811584, | |
| "learning_rate": 5.604554012708873e-05, | |
| "loss": 2.7796, | |
| "step": 353000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.41019803665099974, | |
| "eval_loss": 3.4155313968658447, | |
| "eval_runtime": 154.2799, | |
| "eval_samples_per_second": 375.402, | |
| "eval_steps_per_second": 5.866, | |
| "step": 353324 | |
| }, | |
| { | |
| "epoch": 19.04, | |
| "grad_norm": 0.29620590806007385, | |
| "learning_rate": 5.310367145210638e-05, | |
| "loss": 2.7661, | |
| "step": 354000 | |
| }, | |
| { | |
| "epoch": 19.09, | |
| "grad_norm": 0.3027678430080414, | |
| "learning_rate": 5.016180277712403e-05, | |
| "loss": 2.7561, | |
| "step": 355000 | |
| }, | |
| { | |
| "epoch": 19.14, | |
| "grad_norm": 0.31339025497436523, | |
| "learning_rate": 4.721993410214168e-05, | |
| "loss": 2.757, | |
| "step": 356000 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "grad_norm": 0.31296485662460327, | |
| "learning_rate": 4.4278065427159335e-05, | |
| "loss": 2.7561, | |
| "step": 357000 | |
| }, | |
| { | |
| "epoch": 19.25, | |
| "grad_norm": 0.3046360909938812, | |
| "learning_rate": 4.1339138620851965e-05, | |
| "loss": 2.7601, | |
| "step": 358000 | |
| }, | |
| { | |
| "epoch": 19.31, | |
| "grad_norm": 0.3128166198730469, | |
| "learning_rate": 3.839726994586962e-05, | |
| "loss": 2.7618, | |
| "step": 359000 | |
| }, | |
| { | |
| "epoch": 19.36, | |
| "grad_norm": 0.3523208200931549, | |
| "learning_rate": 3.545540127088727e-05, | |
| "loss": 2.7576, | |
| "step": 360000 | |
| }, | |
| { | |
| "epoch": 19.41, | |
| "grad_norm": 0.30089735984802246, | |
| "learning_rate": 3.25164744645799e-05, | |
| "loss": 2.7602, | |
| "step": 361000 | |
| }, | |
| { | |
| "epoch": 19.47, | |
| "grad_norm": 0.3562370538711548, | |
| "learning_rate": 2.957460578959755e-05, | |
| "loss": 2.7599, | |
| "step": 362000 | |
| }, | |
| { | |
| "epoch": 19.52, | |
| "grad_norm": 0.33757689595222473, | |
| "learning_rate": 2.6635678983290187e-05, | |
| "loss": 2.7591, | |
| "step": 363000 | |
| }, | |
| { | |
| "epoch": 19.57, | |
| "grad_norm": 0.313703715801239, | |
| "learning_rate": 2.3693810308307837e-05, | |
| "loss": 2.7621, | |
| "step": 364000 | |
| }, | |
| { | |
| "epoch": 19.63, | |
| "grad_norm": 0.31429997086524963, | |
| "learning_rate": 2.075488350200047e-05, | |
| "loss": 2.7585, | |
| "step": 365000 | |
| }, | |
| { | |
| "epoch": 19.68, | |
| "grad_norm": 0.29419606924057007, | |
| "learning_rate": 1.7813014827018123e-05, | |
| "loss": 2.7598, | |
| "step": 366000 | |
| }, | |
| { | |
| "epoch": 19.74, | |
| "grad_norm": 0.28869205713272095, | |
| "learning_rate": 1.4874088020710754e-05, | |
| "loss": 2.7631, | |
| "step": 367000 | |
| }, | |
| { | |
| "epoch": 19.79, | |
| "grad_norm": 0.2999878227710724, | |
| "learning_rate": 1.1932219345728408e-05, | |
| "loss": 2.7615, | |
| "step": 368000 | |
| }, | |
| { | |
| "epoch": 19.84, | |
| "grad_norm": 0.31875109672546387, | |
| "learning_rate": 8.990350670746058e-06, | |
| "loss": 2.7607, | |
| "step": 369000 | |
| }, | |
| { | |
| "epoch": 19.9, | |
| "grad_norm": 0.2959778308868408, | |
| "learning_rate": 6.051423864438692e-06, | |
| "loss": 2.7631, | |
| "step": 370000 | |
| }, | |
| { | |
| "epoch": 19.95, | |
| "grad_norm": 0.3109806478023529, | |
| "learning_rate": 3.1095551894563426e-06, | |
| "loss": 2.7606, | |
| "step": 371000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.40996444059320364, | |
| "eval_loss": 3.424755573272705, | |
| "eval_runtime": 154.1805, | |
| "eval_samples_per_second": 375.644, | |
| "eval_steps_per_second": 5.87, | |
| "step": 371920 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 371920, | |
| "total_flos": 1.5670047538944e+18, | |
| "train_loss": 3.0296430246833577, | |
| "train_runtime": 81436.2617, | |
| "train_samples_per_second": 146.143, | |
| "train_steps_per_second": 4.567 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 371920, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 5000, | |
| "total_flos": 1.5670047538944e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |