7035 lines
192 KiB
JSON
7035 lines
192 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.2045706937868958,
|
|
"eval_steps": 500,
|
|
"global_step": 3500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742551231384278,
|
|
"epoch": 0.000292243848266994,
|
|
"grad_norm": 4.9375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7723,
|
|
"mean_token_accuracy": 9.259259095415474e-05,
|
|
"num_tokens": 11114.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742591094970702,
|
|
"epoch": 0.000584487696533988,
|
|
"grad_norm": 5.4375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7466,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 20935.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742585849761962,
|
|
"epoch": 0.000876731544800982,
|
|
"grad_norm": 5.09375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.7483,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 31930.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742614936828613,
|
|
"epoch": 0.001168975393067976,
|
|
"grad_norm": 5.125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.6676,
|
|
"mean_token_accuracy": 0.0005568421445786953,
|
|
"num_tokens": 41672.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742519950866699,
|
|
"epoch": 0.00146121924133497,
|
|
"grad_norm": 4.9375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.5834,
|
|
"mean_token_accuracy": 0.013186294632032514,
|
|
"num_tokens": 52510.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.742103958129883,
|
|
"epoch": 0.001753463089601964,
|
|
"grad_norm": 4.3125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.4654,
|
|
"mean_token_accuracy": 0.04129958841949701,
|
|
"num_tokens": 63214.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.740110874176025,
|
|
"epoch": 0.002045706937868958,
|
|
"grad_norm": 3.203125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.305,
|
|
"mean_token_accuracy": 0.05048500820994377,
|
|
"num_tokens": 73942.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.733905601501466,
|
|
"epoch": 0.002337950786135952,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.1895,
|
|
"mean_token_accuracy": 0.051587145403027536,
|
|
"num_tokens": 84526.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.72324333190918,
|
|
"epoch": 0.0026301946344029457,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.1139,
|
|
"mean_token_accuracy": 0.04853160977363587,
|
|
"num_tokens": 95178.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.714754772186279,
|
|
"epoch": 0.00292243848266994,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.0146,
|
|
"mean_token_accuracy": 0.05312369242310524,
|
|
"num_tokens": 106653.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.708310794830322,
|
|
"epoch": 0.0032146823309369336,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.9113,
|
|
"mean_token_accuracy": 0.050825309753417966,
|
|
"num_tokens": 115969.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.704185581207275,
|
|
"epoch": 0.003506926179203928,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.8939,
|
|
"mean_token_accuracy": 0.048680877685546874,
|
|
"num_tokens": 125826.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.701387882232666,
|
|
"epoch": 0.0037991700274709215,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.872,
|
|
"mean_token_accuracy": 0.04808175079524517,
|
|
"num_tokens": 136677.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.697122478485108,
|
|
"epoch": 0.004091413875737916,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.8236,
|
|
"mean_token_accuracy": 0.04725121222436428,
|
|
"num_tokens": 147535.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.691372871398926,
|
|
"epoch": 0.00438365772400491,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.722,
|
|
"mean_token_accuracy": 0.04985377714037895,
|
|
"num_tokens": 157524.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.681602478027344,
|
|
"epoch": 0.004675901572271904,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.6896,
|
|
"mean_token_accuracy": 0.05114552080631256,
|
|
"num_tokens": 168344.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.67192096710205,
|
|
"epoch": 0.004968145420538897,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.6618,
|
|
"mean_token_accuracy": 0.04409446604549885,
|
|
"num_tokens": 178976.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.664519691467286,
|
|
"epoch": 0.0052603892688058915,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.5639,
|
|
"mean_token_accuracy": 0.05372482612729072,
|
|
"num_tokens": 190567.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.647130489349365,
|
|
"epoch": 0.005552633117072886,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.4984,
|
|
"mean_token_accuracy": 0.05118483640253544,
|
|
"num_tokens": 201587.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.62646837234497,
|
|
"epoch": 0.00584487696533988,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.4193,
|
|
"mean_token_accuracy": 0.04868996068835259,
|
|
"num_tokens": 211913.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.610049438476562,
|
|
"epoch": 0.006137120813606874,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.3409,
|
|
"mean_token_accuracy": 0.05402343980967998,
|
|
"num_tokens": 222247.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.584276962280274,
|
|
"epoch": 0.006429364661873867,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.2142,
|
|
"mean_token_accuracy": 0.06452971324324608,
|
|
"num_tokens": 232361.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.533616924285889,
|
|
"epoch": 0.006721608510140861,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.1281,
|
|
"mean_token_accuracy": 0.0630628488957882,
|
|
"num_tokens": 242885.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.462300109863282,
|
|
"epoch": 0.007013852358407856,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 9.0436,
|
|
"mean_token_accuracy": 0.060920076444745064,
|
|
"num_tokens": 254482.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.408651351928711,
|
|
"epoch": 0.00730609620667485,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.93,
|
|
"mean_token_accuracy": 0.059300025925040245,
|
|
"num_tokens": 265744.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.367331790924073,
|
|
"epoch": 0.007598340054941843,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.8327,
|
|
"mean_token_accuracy": 0.058924198523163794,
|
|
"num_tokens": 276924.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.272025775909423,
|
|
"epoch": 0.007890583903208837,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.7304,
|
|
"mean_token_accuracy": 0.0622296292334795,
|
|
"num_tokens": 287779.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.183625984191895,
|
|
"epoch": 0.008182827751475831,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.6335,
|
|
"mean_token_accuracy": 0.06356315091252326,
|
|
"num_tokens": 298807.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.097103500366211,
|
|
"epoch": 0.008475071599742826,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.5907,
|
|
"mean_token_accuracy": 0.0559815414249897,
|
|
"num_tokens": 309760.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.004411506652833,
|
|
"epoch": 0.00876731544800982,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.397,
|
|
"mean_token_accuracy": 0.06126284077763557,
|
|
"num_tokens": 319750.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.832055950164795,
|
|
"epoch": 0.009059559296276814,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.2924,
|
|
"mean_token_accuracy": 0.0603407833725214,
|
|
"num_tokens": 329931.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.66302547454834,
|
|
"epoch": 0.009351803144543808,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.2478,
|
|
"mean_token_accuracy": 0.06180734224617481,
|
|
"num_tokens": 340457.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.49033498764038,
|
|
"epoch": 0.009644046992810802,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.1199,
|
|
"mean_token_accuracy": 0.06424234174191952,
|
|
"num_tokens": 352450.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.311860942840577,
|
|
"epoch": 0.009936290841077795,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.1159,
|
|
"mean_token_accuracy": 0.06317566409707069,
|
|
"num_tokens": 363204.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.153958892822265,
|
|
"epoch": 0.010228534689344789,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.0882,
|
|
"mean_token_accuracy": 0.062278729677200315,
|
|
"num_tokens": 373577.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.929795932769775,
|
|
"epoch": 0.010520778537611783,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.9663,
|
|
"mean_token_accuracy": 0.07208616435527801,
|
|
"num_tokens": 384267.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.768066787719727,
|
|
"epoch": 0.010813022385878777,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.9324,
|
|
"mean_token_accuracy": 0.06511378176510334,
|
|
"num_tokens": 394767.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.666393280029297,
|
|
"epoch": 0.011105266234145771,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.9172,
|
|
"mean_token_accuracy": 0.07024285569787025,
|
|
"num_tokens": 405807.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.585619068145752,
|
|
"epoch": 0.011397510082412765,
|
|
"grad_norm": 0.68359375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.869,
|
|
"mean_token_accuracy": 0.0643521461635828,
|
|
"num_tokens": 416381.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.565132427215577,
|
|
"epoch": 0.01168975393067976,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.9632,
|
|
"mean_token_accuracy": 0.06446416340768338,
|
|
"num_tokens": 427262.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.418638515472413,
|
|
"epoch": 0.011981997778946754,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.8962,
|
|
"mean_token_accuracy": 0.06822493113577366,
|
|
"num_tokens": 438504.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.47780933380127,
|
|
"epoch": 0.012274241627213748,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.8873,
|
|
"mean_token_accuracy": 0.06689661256968975,
|
|
"num_tokens": 449923.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.430081558227538,
|
|
"epoch": 0.01256648547548074,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.8463,
|
|
"mean_token_accuracy": 0.0716497365385294,
|
|
"num_tokens": 460736.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.367777156829835,
|
|
"epoch": 0.012858729323747735,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.9365,
|
|
"mean_token_accuracy": 0.0605232123285532,
|
|
"num_tokens": 471513.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.417449378967286,
|
|
"epoch": 0.013150973172014729,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.7879,
|
|
"mean_token_accuracy": 0.07277968265116215,
|
|
"num_tokens": 482115.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.359326553344726,
|
|
"epoch": 0.013443217020281723,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.8079,
|
|
"mean_token_accuracy": 0.07002997919917106,
|
|
"num_tokens": 493273.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.321810531616212,
|
|
"epoch": 0.013735460868548717,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.8188,
|
|
"mean_token_accuracy": 0.07201721332967281,
|
|
"num_tokens": 503939.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.311158466339112,
|
|
"epoch": 0.014027704716815711,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.7369,
|
|
"mean_token_accuracy": 0.06772977598011494,
|
|
"num_tokens": 515215.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.281570434570312,
|
|
"epoch": 0.014319948565082705,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.8223,
|
|
"mean_token_accuracy": 0.07254013754427432,
|
|
"num_tokens": 526485.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.277181148529053,
|
|
"epoch": 0.0146121924133497,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.7672,
|
|
"mean_token_accuracy": 0.06958542391657829,
|
|
"num_tokens": 536735.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.277847576141358,
|
|
"epoch": 0.014904436261616694,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.8541,
|
|
"mean_token_accuracy": 0.06499786637723445,
|
|
"num_tokens": 547326.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.180327033996582,
|
|
"epoch": 0.015196680109883686,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.7502,
|
|
"mean_token_accuracy": 0.06869623251259327,
|
|
"num_tokens": 558275.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.257558250427246,
|
|
"epoch": 0.01548892395815068,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.7096,
|
|
"mean_token_accuracy": 0.07516702339053154,
|
|
"num_tokens": 569711.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.150344276428223,
|
|
"epoch": 0.015781167806417674,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.6881,
|
|
"mean_token_accuracy": 0.07785917520523071,
|
|
"num_tokens": 579602.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.24105463027954,
|
|
"epoch": 0.01607341165468467,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.7286,
|
|
"mean_token_accuracy": 0.07401030212640762,
|
|
"num_tokens": 590899.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.173699569702148,
|
|
"epoch": 0.016365655502951663,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.726,
|
|
"mean_token_accuracy": 0.06962668001651764,
|
|
"num_tokens": 602080.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.16687994003296,
|
|
"epoch": 0.016657899351218655,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.7605,
|
|
"mean_token_accuracy": 0.07070777006447315,
|
|
"num_tokens": 613692.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.145956993103027,
|
|
"epoch": 0.01695014319948565,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.7305,
|
|
"mean_token_accuracy": 0.07069781571626663,
|
|
"num_tokens": 625019.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.123618793487548,
|
|
"epoch": 0.017242387047752643,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.6456,
|
|
"mean_token_accuracy": 0.07288045659661294,
|
|
"num_tokens": 636453.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.108728456497193,
|
|
"epoch": 0.01753463089601964,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.6918,
|
|
"mean_token_accuracy": 0.07240097671747207,
|
|
"num_tokens": 647272.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.040448427200317,
|
|
"epoch": 0.017826874744286632,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.701,
|
|
"mean_token_accuracy": 0.07080473527312278,
|
|
"num_tokens": 657724.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.166200351715087,
|
|
"epoch": 0.018119118592553628,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.7053,
|
|
"mean_token_accuracy": 0.07137923240661621,
|
|
"num_tokens": 669183.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.084780406951904,
|
|
"epoch": 0.01841136244082062,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.6679,
|
|
"mean_token_accuracy": 0.06901021562516689,
|
|
"num_tokens": 680439.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.157064056396484,
|
|
"epoch": 0.018703606289087616,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.712,
|
|
"mean_token_accuracy": 0.0706634670495987,
|
|
"num_tokens": 690789.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.046548128128052,
|
|
"epoch": 0.01899585013735461,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.6831,
|
|
"mean_token_accuracy": 0.07187991701066494,
|
|
"num_tokens": 701818.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.062455224990845,
|
|
"epoch": 0.019288093985621604,
|
|
"grad_norm": 0.71484375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.6316,
|
|
"mean_token_accuracy": 0.07437365353107453,
|
|
"num_tokens": 713344.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.084509325027465,
|
|
"epoch": 0.019580337833888597,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.6384,
|
|
"mean_token_accuracy": 0.07300028279423713,
|
|
"num_tokens": 723279.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.018547964096069,
|
|
"epoch": 0.01987258168215559,
|
|
"grad_norm": 0.73046875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.625,
|
|
"mean_token_accuracy": 0.07260205298662185,
|
|
"num_tokens": 733875.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.072830390930175,
|
|
"epoch": 0.020164825530422585,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.6669,
|
|
"mean_token_accuracy": 0.0713542215526104,
|
|
"num_tokens": 745449.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.05644793510437,
|
|
"epoch": 0.020457069378689578,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.6741,
|
|
"mean_token_accuracy": 0.0740118071436882,
|
|
"num_tokens": 757570.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.017089939117431,
|
|
"epoch": 0.020749313226956573,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.5737,
|
|
"mean_token_accuracy": 0.08025477081537247,
|
|
"num_tokens": 767601.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.00361680984497,
|
|
"epoch": 0.021041557075223566,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.6323,
|
|
"mean_token_accuracy": 0.0752041395753622,
|
|
"num_tokens": 777852.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.018228721618652,
|
|
"epoch": 0.021333800923490562,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.5859,
|
|
"mean_token_accuracy": 0.0751498755067587,
|
|
"num_tokens": 788339.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.970228672027588,
|
|
"epoch": 0.021626044771757554,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.6588,
|
|
"mean_token_accuracy": 0.07353257052600384,
|
|
"num_tokens": 799791.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.015653133392334,
|
|
"epoch": 0.02191828862002455,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.5746,
|
|
"mean_token_accuracy": 0.07374845892190933,
|
|
"num_tokens": 810141.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.040093088150025,
|
|
"epoch": 0.022210532468291543,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.6182,
|
|
"mean_token_accuracy": 0.07429262697696686,
|
|
"num_tokens": 820929.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.9526266098022464,
|
|
"epoch": 0.022502776316558535,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.5519,
|
|
"mean_token_accuracy": 0.07891798727214336,
|
|
"num_tokens": 831340.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.935062026977539,
|
|
"epoch": 0.02279502016482553,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.5636,
|
|
"mean_token_accuracy": 0.0735638827085495,
|
|
"num_tokens": 841846.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.029356002807617,
|
|
"epoch": 0.023087264013092523,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.5434,
|
|
"mean_token_accuracy": 0.07343484573066235,
|
|
"num_tokens": 852089.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.838059997558593,
|
|
"epoch": 0.02337950786135952,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.5241,
|
|
"mean_token_accuracy": 0.07688843682408333,
|
|
"num_tokens": 862568.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.034818124771117,
|
|
"epoch": 0.02367175170962651,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.5905,
|
|
"mean_token_accuracy": 0.07528307065367698,
|
|
"num_tokens": 874676.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.867450904846192,
|
|
"epoch": 0.023963995557893508,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.5115,
|
|
"mean_token_accuracy": 0.08100188933312893,
|
|
"num_tokens": 886338.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.946144676208496,
|
|
"epoch": 0.0242562394061605,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.5287,
|
|
"mean_token_accuracy": 0.07384060397744178,
|
|
"num_tokens": 896671.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.923208522796631,
|
|
"epoch": 0.024548483254427496,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.5025,
|
|
"mean_token_accuracy": 0.07231347896158695,
|
|
"num_tokens": 908265.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.851961994171143,
|
|
"epoch": 0.02484072710269449,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.4692,
|
|
"mean_token_accuracy": 0.07885380312800408,
|
|
"num_tokens": 919024.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.894201850891113,
|
|
"epoch": 0.02513297095096148,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.4639,
|
|
"mean_token_accuracy": 0.08095669150352477,
|
|
"num_tokens": 930837.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.845162343978882,
|
|
"epoch": 0.025425214799228477,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.4312,
|
|
"mean_token_accuracy": 0.076503673940897,
|
|
"num_tokens": 941374.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.838292598724365,
|
|
"epoch": 0.02571745864749547,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.3796,
|
|
"mean_token_accuracy": 0.0836145281791687,
|
|
"num_tokens": 953273.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.836203670501709,
|
|
"epoch": 0.026009702495762465,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.3659,
|
|
"mean_token_accuracy": 0.09230264946818352,
|
|
"num_tokens": 964882.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.860473203659057,
|
|
"epoch": 0.026301946344029457,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.5208,
|
|
"mean_token_accuracy": 0.08153782263398171,
|
|
"num_tokens": 974732.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.774071216583252,
|
|
"epoch": 0.026594190192296453,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.4513,
|
|
"mean_token_accuracy": 0.07568450495600701,
|
|
"num_tokens": 985709.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.999471426010132,
|
|
"epoch": 0.026886434040563446,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.4197,
|
|
"mean_token_accuracy": 0.07974797300994396,
|
|
"num_tokens": 998177.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.711365842819214,
|
|
"epoch": 0.02717867788883044,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.3536,
|
|
"mean_token_accuracy": 0.08135458827018738,
|
|
"num_tokens": 1008676.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.800386238098144,
|
|
"epoch": 0.027470921737097434,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.4651,
|
|
"mean_token_accuracy": 0.07976322658360005,
|
|
"num_tokens": 1019951.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.785122346878052,
|
|
"epoch": 0.027763165585364426,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.3981,
|
|
"mean_token_accuracy": 0.0761278249323368,
|
|
"num_tokens": 1030219.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.781033182144165,
|
|
"epoch": 0.028055409433631422,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.4141,
|
|
"mean_token_accuracy": 0.07795999571681023,
|
|
"num_tokens": 1041911.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.891617631912231,
|
|
"epoch": 0.028347653281898415,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.4312,
|
|
"mean_token_accuracy": 0.08173990175127983,
|
|
"num_tokens": 1052638.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.774001789093018,
|
|
"epoch": 0.02863989713016541,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.3755,
|
|
"mean_token_accuracy": 0.08225451484322548,
|
|
"num_tokens": 1063465.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.861321496963501,
|
|
"epoch": 0.028932140978432403,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.3888,
|
|
"mean_token_accuracy": 0.08350497037172318,
|
|
"num_tokens": 1073583.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.868850564956665,
|
|
"epoch": 0.0292243848266994,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.4377,
|
|
"mean_token_accuracy": 0.08040728494524955,
|
|
"num_tokens": 1083961.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.663293409347534,
|
|
"epoch": 0.02951662867496639,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.3053,
|
|
"mean_token_accuracy": 0.08032148070633412,
|
|
"num_tokens": 1095339.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.742640733718872,
|
|
"epoch": 0.029808872523233387,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.3746,
|
|
"mean_token_accuracy": 0.07911042794585228,
|
|
"num_tokens": 1105518.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.822371339797973,
|
|
"epoch": 0.03010111637150038,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.2683,
|
|
"mean_token_accuracy": 0.08846351876854897,
|
|
"num_tokens": 1115666.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.7058703899383545,
|
|
"epoch": 0.030393360219767372,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.3901,
|
|
"mean_token_accuracy": 0.08510988727211952,
|
|
"num_tokens": 1125827.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.762892246246338,
|
|
"epoch": 0.030685604068034368,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.3453,
|
|
"mean_token_accuracy": 0.08135025277733803,
|
|
"num_tokens": 1136154.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.712664318084717,
|
|
"epoch": 0.03097784791630136,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.3361,
|
|
"mean_token_accuracy": 0.08622926697134972,
|
|
"num_tokens": 1146392.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.7614504337310795,
|
|
"epoch": 0.03127009176456835,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.3515,
|
|
"mean_token_accuracy": 0.08140693604946136,
|
|
"num_tokens": 1157566.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.6518913269042965,
|
|
"epoch": 0.03156233561283535,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.3194,
|
|
"mean_token_accuracy": 0.0838658906519413,
|
|
"num_tokens": 1167795.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.775746774673462,
|
|
"epoch": 0.031854579461102345,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.2691,
|
|
"mean_token_accuracy": 0.09419989511370659,
|
|
"num_tokens": 1178894.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.634786605834961,
|
|
"epoch": 0.03214682330936934,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.2696,
|
|
"mean_token_accuracy": 0.08667804896831513,
|
|
"num_tokens": 1189361.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.752673006057739,
|
|
"epoch": 0.03243906715763633,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.3135,
|
|
"mean_token_accuracy": 0.08304268047213555,
|
|
"num_tokens": 1198992.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.692470407485962,
|
|
"epoch": 0.032731311005903325,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.294,
|
|
"mean_token_accuracy": 0.07943044230341911,
|
|
"num_tokens": 1208732.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.655223751068116,
|
|
"epoch": 0.03302355485417032,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.2954,
|
|
"mean_token_accuracy": 0.08259731158614159,
|
|
"num_tokens": 1219015.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.612499952316284,
|
|
"epoch": 0.03331579870243731,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.1795,
|
|
"mean_token_accuracy": 0.1000419594347477,
|
|
"num_tokens": 1229227.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.750643157958985,
|
|
"epoch": 0.033608042550704306,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.3098,
|
|
"mean_token_accuracy": 0.08021004945039749,
|
|
"num_tokens": 1239654.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.618366050720215,
|
|
"epoch": 0.0339002863989713,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.3005,
|
|
"mean_token_accuracy": 0.08035954535007477,
|
|
"num_tokens": 1250955.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.729614543914795,
|
|
"epoch": 0.0341925302472383,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.2497,
|
|
"mean_token_accuracy": 0.08898582234978676,
|
|
"num_tokens": 1262574.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.672766494750976,
|
|
"epoch": 0.03448477409550529,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.2059,
|
|
"mean_token_accuracy": 0.08972268775105477,
|
|
"num_tokens": 1273272.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.505560779571534,
|
|
"epoch": 0.03477701794377228,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.1898,
|
|
"mean_token_accuracy": 0.09178911671042442,
|
|
"num_tokens": 1284001.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.633696794509888,
|
|
"epoch": 0.03506926179203928,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.2354,
|
|
"mean_token_accuracy": 0.08889753520488738,
|
|
"num_tokens": 1294671.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.695655345916748,
|
|
"epoch": 0.035361505640306275,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.2349,
|
|
"mean_token_accuracy": 0.09624799937009812,
|
|
"num_tokens": 1305397.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.568394422531128,
|
|
"epoch": 0.035653749488573264,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.2377,
|
|
"mean_token_accuracy": 0.08420243933796882,
|
|
"num_tokens": 1317449.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.562969303131103,
|
|
"epoch": 0.03594599333684026,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.176,
|
|
"mean_token_accuracy": 0.08269712552428246,
|
|
"num_tokens": 1328645.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.559274625778198,
|
|
"epoch": 0.036238237185107255,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.1652,
|
|
"mean_token_accuracy": 0.09311013892292977,
|
|
"num_tokens": 1339815.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.583710241317749,
|
|
"epoch": 0.036530481033374244,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.1795,
|
|
"mean_token_accuracy": 0.08754727616906166,
|
|
"num_tokens": 1350301.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.535983610153198,
|
|
"epoch": 0.03682272488164124,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.1724,
|
|
"mean_token_accuracy": 0.08950133547186852,
|
|
"num_tokens": 1360928.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.625943231582641,
|
|
"epoch": 0.037114968729908236,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.1957,
|
|
"mean_token_accuracy": 0.08902273774147033,
|
|
"num_tokens": 1371618.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.511281824111938,
|
|
"epoch": 0.03740721257817523,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.1562,
|
|
"mean_token_accuracy": 0.08582355380058289,
|
|
"num_tokens": 1382824.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.470151281356811,
|
|
"epoch": 0.03769945642644222,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.1191,
|
|
"mean_token_accuracy": 0.09281812384724616,
|
|
"num_tokens": 1393244.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.475745964050293,
|
|
"epoch": 0.03799170027470922,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.0909,
|
|
"mean_token_accuracy": 0.09678038656711578,
|
|
"num_tokens": 1403469.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.615570402145385,
|
|
"epoch": 0.03828394412297621,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.1651,
|
|
"mean_token_accuracy": 0.09157615751028061,
|
|
"num_tokens": 1413860.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.553770732879639,
|
|
"epoch": 0.03857618797124321,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.2248,
|
|
"mean_token_accuracy": 0.08392854556441307,
|
|
"num_tokens": 1423985.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.531700229644775,
|
|
"epoch": 0.0388684318195102,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.1654,
|
|
"mean_token_accuracy": 0.09291392266750335,
|
|
"num_tokens": 1434838.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.482819223403931,
|
|
"epoch": 0.039160675667777194,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.0905,
|
|
"mean_token_accuracy": 0.09290852397680283,
|
|
"num_tokens": 1446429.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.611834764480591,
|
|
"epoch": 0.03945291951604419,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.1879,
|
|
"mean_token_accuracy": 0.08893333822488785,
|
|
"num_tokens": 1456726.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.5248387336730955,
|
|
"epoch": 0.03974516336431118,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.0781,
|
|
"mean_token_accuracy": 0.09758968204259873,
|
|
"num_tokens": 1466378.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.323547267913819,
|
|
"epoch": 0.040037407212578174,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.0345,
|
|
"mean_token_accuracy": 0.09719507321715355,
|
|
"num_tokens": 1477755.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.507241344451904,
|
|
"epoch": 0.04032965106084517,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.1006,
|
|
"mean_token_accuracy": 0.09118306115269662,
|
|
"num_tokens": 1488209.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.419371461868286,
|
|
"epoch": 0.040621894909112166,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.1465,
|
|
"mean_token_accuracy": 0.08755657598376274,
|
|
"num_tokens": 1498658.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.546753549575806,
|
|
"epoch": 0.040914138757379155,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.1576,
|
|
"mean_token_accuracy": 0.08506937101483344,
|
|
"num_tokens": 1509766.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.439605093002319,
|
|
"epoch": 0.04120638260564615,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.036,
|
|
"mean_token_accuracy": 0.0939096562564373,
|
|
"num_tokens": 1521139.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.4743701934814455,
|
|
"epoch": 0.04149862645391315,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.1115,
|
|
"mean_token_accuracy": 0.08748833164572715,
|
|
"num_tokens": 1531875.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.395281553268433,
|
|
"epoch": 0.041790870302180136,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.0801,
|
|
"mean_token_accuracy": 0.08955910429358482,
|
|
"num_tokens": 1542678.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.462686967849732,
|
|
"epoch": 0.04208311415044713,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.0268,
|
|
"mean_token_accuracy": 0.09358467385172844,
|
|
"num_tokens": 1553882.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.43300313949585,
|
|
"epoch": 0.04237535799871413,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.1176,
|
|
"mean_token_accuracy": 0.08562448695302009,
|
|
"num_tokens": 1564854.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.425636720657349,
|
|
"epoch": 0.042667601846981124,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.0026,
|
|
"mean_token_accuracy": 0.09751182347536087,
|
|
"num_tokens": 1576112.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.374439096450805,
|
|
"epoch": 0.04295984569524811,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.0378,
|
|
"mean_token_accuracy": 0.09424405097961426,
|
|
"num_tokens": 1586235.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.347483777999878,
|
|
"epoch": 0.04325208954351511,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.0347,
|
|
"mean_token_accuracy": 0.0911446213722229,
|
|
"num_tokens": 1596727.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.407256555557251,
|
|
"epoch": 0.043544333391782104,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.9573,
|
|
"mean_token_accuracy": 0.09358611106872558,
|
|
"num_tokens": 1607008.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.389186716079712,
|
|
"epoch": 0.0438365772400491,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.0579,
|
|
"mean_token_accuracy": 0.0928824670612812,
|
|
"num_tokens": 1618081.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.385285234451294,
|
|
"epoch": 0.04412882108831609,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.0165,
|
|
"mean_token_accuracy": 0.08930294588208199,
|
|
"num_tokens": 1629095.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.394097995758057,
|
|
"epoch": 0.044421064936583085,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.0118,
|
|
"mean_token_accuracy": 0.093733761459589,
|
|
"num_tokens": 1639861.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.372588682174682,
|
|
"epoch": 0.04471330878485008,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.0534,
|
|
"mean_token_accuracy": 0.09636619538068772,
|
|
"num_tokens": 1649169.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.278450775146484,
|
|
"epoch": 0.04500555263311707,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.9357,
|
|
"mean_token_accuracy": 0.1010350152850151,
|
|
"num_tokens": 1658528.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.303726387023926,
|
|
"epoch": 0.045297796481384066,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.0068,
|
|
"mean_token_accuracy": 0.09450188651680946,
|
|
"num_tokens": 1669231.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.309107494354248,
|
|
"epoch": 0.04559004032965106,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.9475,
|
|
"mean_token_accuracy": 0.09322728961706161,
|
|
"num_tokens": 1680503.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.336097192764282,
|
|
"epoch": 0.04588228417791806,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.9766,
|
|
"mean_token_accuracy": 0.09063231274485588,
|
|
"num_tokens": 1690827.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.370522546768188,
|
|
"epoch": 0.04617452802618505,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.0418,
|
|
"mean_token_accuracy": 0.09461386948823929,
|
|
"num_tokens": 1702931.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.289389944076538,
|
|
"epoch": 0.04646677187445204,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.9307,
|
|
"mean_token_accuracy": 0.09922353029251099,
|
|
"num_tokens": 1712834.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.3177947998046875,
|
|
"epoch": 0.04675901572271904,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.0006,
|
|
"mean_token_accuracy": 0.0919931598007679,
|
|
"num_tokens": 1723289.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.245712804794311,
|
|
"epoch": 0.04705125957098603,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.9702,
|
|
"mean_token_accuracy": 0.09567693397402763,
|
|
"num_tokens": 1734477.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.306273651123047,
|
|
"epoch": 0.04734350341925302,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.9762,
|
|
"mean_token_accuracy": 0.09110540002584458,
|
|
"num_tokens": 1745187.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.276857900619507,
|
|
"epoch": 0.04763574726752002,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.9011,
|
|
"mean_token_accuracy": 0.10115249156951904,
|
|
"num_tokens": 1756687.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.304260778427124,
|
|
"epoch": 0.047927991115787015,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.024,
|
|
"mean_token_accuracy": 0.09978155121207237,
|
|
"num_tokens": 1767928.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.309319734573364,
|
|
"epoch": 0.048220234964054004,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.9585,
|
|
"mean_token_accuracy": 0.09580064937472343,
|
|
"num_tokens": 1777951.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.194837522506714,
|
|
"epoch": 0.048512478812321,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.9514,
|
|
"mean_token_accuracy": 0.09750414118170739,
|
|
"num_tokens": 1788598.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.25841178894043,
|
|
"epoch": 0.048804722660587996,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.9107,
|
|
"mean_token_accuracy": 0.10313281938433647,
|
|
"num_tokens": 1799446.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.308444929122925,
|
|
"epoch": 0.04909696650885499,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.93,
|
|
"mean_token_accuracy": 0.09639307036995888,
|
|
"num_tokens": 1809687.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.247332191467285,
|
|
"epoch": 0.04938921035712198,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.9496,
|
|
"mean_token_accuracy": 0.10043755695223808,
|
|
"num_tokens": 1820535.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.192328262329101,
|
|
"epoch": 0.04968145420538898,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.8901,
|
|
"mean_token_accuracy": 0.09903936088085175,
|
|
"num_tokens": 1830834.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.244427633285523,
|
|
"epoch": 0.04997369805365597,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.8568,
|
|
"mean_token_accuracy": 0.10095973312854767,
|
|
"num_tokens": 1842615.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.1798933982849125,
|
|
"epoch": 0.05026594190192296,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.8778,
|
|
"mean_token_accuracy": 0.10051687061786652,
|
|
"num_tokens": 1852844.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.210625696182251,
|
|
"epoch": 0.05055818575018996,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.9139,
|
|
"mean_token_accuracy": 0.09776052087545395,
|
|
"num_tokens": 1862528.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.177963972091675,
|
|
"epoch": 0.05085042959845695,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.8757,
|
|
"mean_token_accuracy": 0.1031526930630207,
|
|
"num_tokens": 1873275.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.275309658050537,
|
|
"epoch": 0.05114267344672395,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.9153,
|
|
"mean_token_accuracy": 0.10134659111499786,
|
|
"num_tokens": 1884724.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.248415899276734,
|
|
"epoch": 0.05143491729499094,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.8861,
|
|
"mean_token_accuracy": 0.09837430268526078,
|
|
"num_tokens": 1894705.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.133459377288818,
|
|
"epoch": 0.051727161143257934,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.9485,
|
|
"mean_token_accuracy": 0.09655695110559463,
|
|
"num_tokens": 1906553.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.257501792907715,
|
|
"epoch": 0.05201940499152493,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.8849,
|
|
"mean_token_accuracy": 0.1023280717432499,
|
|
"num_tokens": 1916628.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.0709809303283695,
|
|
"epoch": 0.052311648839791926,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.725,
|
|
"mean_token_accuracy": 0.11589453369379044,
|
|
"num_tokens": 1926686.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.150864934921264,
|
|
"epoch": 0.052603892688058915,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.9068,
|
|
"mean_token_accuracy": 0.101015555113554,
|
|
"num_tokens": 1938150.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.1812304019927975,
|
|
"epoch": 0.05289613653632591,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.8918,
|
|
"mean_token_accuracy": 0.097091793268919,
|
|
"num_tokens": 1948617.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.160140323638916,
|
|
"epoch": 0.053188380384592907,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.8934,
|
|
"mean_token_accuracy": 0.10218235999345779,
|
|
"num_tokens": 1958943.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.165960168838501,
|
|
"epoch": 0.053480624232859895,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.8192,
|
|
"mean_token_accuracy": 0.10898060277104378,
|
|
"num_tokens": 1968988.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.051486253738403,
|
|
"epoch": 0.05377286808112689,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.7949,
|
|
"mean_token_accuracy": 0.10556035414338112,
|
|
"num_tokens": 1979490.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.28261432647705,
|
|
"epoch": 0.05406511192939389,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.8574,
|
|
"mean_token_accuracy": 0.10469735637307168,
|
|
"num_tokens": 1990377.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.070431137084961,
|
|
"epoch": 0.05435735577766088,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.8361,
|
|
"mean_token_accuracy": 0.10013288781046867,
|
|
"num_tokens": 2001678.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.103927278518677,
|
|
"epoch": 0.05464959962592787,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.8266,
|
|
"mean_token_accuracy": 0.10559705421328544,
|
|
"num_tokens": 2011550.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.146097993850708,
|
|
"epoch": 0.05494184347419487,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.8591,
|
|
"mean_token_accuracy": 0.10166500806808472,
|
|
"num_tokens": 2021607.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.074561977386475,
|
|
"epoch": 0.055234087322461864,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.831,
|
|
"mean_token_accuracy": 0.10222307071089745,
|
|
"num_tokens": 2030975.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.097246503829956,
|
|
"epoch": 0.05552633117072885,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.9034,
|
|
"mean_token_accuracy": 0.09934694394469261,
|
|
"num_tokens": 2041799.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.116580200195313,
|
|
"epoch": 0.05581857501899585,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.8168,
|
|
"mean_token_accuracy": 0.10345752537250519,
|
|
"num_tokens": 2052960.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.11494288444519,
|
|
"epoch": 0.056110818867262845,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.8662,
|
|
"mean_token_accuracy": 0.09989891946315765,
|
|
"num_tokens": 2064156.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.182250165939331,
|
|
"epoch": 0.05640306271552984,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.8254,
|
|
"mean_token_accuracy": 0.10051817372441292,
|
|
"num_tokens": 2074774.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.077956199645996,
|
|
"epoch": 0.05669530656379683,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.8527,
|
|
"mean_token_accuracy": 0.10035398751497268,
|
|
"num_tokens": 2086164.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.885358047485352,
|
|
"epoch": 0.056987550412063825,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.672,
|
|
"mean_token_accuracy": 0.127625822275877,
|
|
"num_tokens": 2096268.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.233099365234375,
|
|
"epoch": 0.05727979426033082,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.8813,
|
|
"mean_token_accuracy": 0.09680663272738457,
|
|
"num_tokens": 2108511.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.136028242111206,
|
|
"epoch": 0.05757203810859782,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.8085,
|
|
"mean_token_accuracy": 0.0988966628909111,
|
|
"num_tokens": 2119622.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.998387289047241,
|
|
"epoch": 0.057864281956864806,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.8613,
|
|
"mean_token_accuracy": 0.10771243944764138,
|
|
"num_tokens": 2131021.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.106973361968994,
|
|
"epoch": 0.0581565258051318,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.8496,
|
|
"mean_token_accuracy": 0.10358692184090615,
|
|
"num_tokens": 2141797.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.108656787872315,
|
|
"epoch": 0.0584487696533988,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.7338,
|
|
"mean_token_accuracy": 0.10600685104727745,
|
|
"num_tokens": 2153609.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.041635751724243,
|
|
"epoch": 0.05874101350166579,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.8544,
|
|
"mean_token_accuracy": 0.10038745403289795,
|
|
"num_tokens": 2164237.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.159450626373291,
|
|
"epoch": 0.05903325734993278,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.8092,
|
|
"mean_token_accuracy": 0.09949537962675095,
|
|
"num_tokens": 2175035.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.941095781326294,
|
|
"epoch": 0.05932550119819978,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.7273,
|
|
"mean_token_accuracy": 0.10634237006306649,
|
|
"num_tokens": 2186217.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.0672167301177975,
|
|
"epoch": 0.059617745046466775,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.777,
|
|
"mean_token_accuracy": 0.10704884603619576,
|
|
"num_tokens": 2197290.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.999376821517944,
|
|
"epoch": 0.059909988894733764,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.8155,
|
|
"mean_token_accuracy": 0.0990687295794487,
|
|
"num_tokens": 2208835.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.138485765457153,
|
|
"epoch": 0.06020223274300076,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.7249,
|
|
"mean_token_accuracy": 0.10986540690064431,
|
|
"num_tokens": 2219153.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.9783261775970455,
|
|
"epoch": 0.060494476591267755,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.7549,
|
|
"mean_token_accuracy": 0.10616105198860168,
|
|
"num_tokens": 2230289.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.989655160903931,
|
|
"epoch": 0.060786720439534744,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 6.8395,
|
|
"mean_token_accuracy": 0.10555440336465835,
|
|
"num_tokens": 2241768.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.044928169250488,
|
|
"epoch": 0.06107896428780174,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 6.7114,
|
|
"mean_token_accuracy": 0.11182207837700844,
|
|
"num_tokens": 2253249.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.989438867568969,
|
|
"epoch": 0.061371208136068736,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.7792,
|
|
"mean_token_accuracy": 0.09858244881033898,
|
|
"num_tokens": 2263961.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.941675472259521,
|
|
"epoch": 0.06166345198433573,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.6945,
|
|
"mean_token_accuracy": 0.11112689971923828,
|
|
"num_tokens": 2274016.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.96781735420227,
|
|
"epoch": 0.06195569583260272,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.7391,
|
|
"mean_token_accuracy": 0.10291829779744148,
|
|
"num_tokens": 2285164.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.105303335189819,
|
|
"epoch": 0.06224793968086972,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.7238,
|
|
"mean_token_accuracy": 0.10441325902938843,
|
|
"num_tokens": 2294887.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.961658096313476,
|
|
"epoch": 0.0625401835291367,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.7424,
|
|
"mean_token_accuracy": 0.10544760972261429,
|
|
"num_tokens": 2305787.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.00815224647522,
|
|
"epoch": 0.0628324273774037,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.7683,
|
|
"mean_token_accuracy": 0.10668534934520721,
|
|
"num_tokens": 2316756.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.019680690765381,
|
|
"epoch": 0.0631246712256707,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.783,
|
|
"mean_token_accuracy": 0.10868853554129601,
|
|
"num_tokens": 2327335.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.018626260757446,
|
|
"epoch": 0.0634169150739377,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.7444,
|
|
"mean_token_accuracy": 0.1080512523651123,
|
|
"num_tokens": 2337182.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.036516618728638,
|
|
"epoch": 0.06370915892220469,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.7131,
|
|
"mean_token_accuracy": 0.10611762776970864,
|
|
"num_tokens": 2347843.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.903224134445191,
|
|
"epoch": 0.06400140277047169,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.7136,
|
|
"mean_token_accuracy": 0.10907531604170799,
|
|
"num_tokens": 2358159.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.897138977050782,
|
|
"epoch": 0.06429364661873868,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.6937,
|
|
"mean_token_accuracy": 0.10973558053374291,
|
|
"num_tokens": 2368191.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.996454906463623,
|
|
"epoch": 0.06458589046700566,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.6386,
|
|
"mean_token_accuracy": 0.11236895397305488,
|
|
"num_tokens": 2377672.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.894516134262085,
|
|
"epoch": 0.06487813431527266,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.7148,
|
|
"mean_token_accuracy": 0.11475707441568375,
|
|
"num_tokens": 2387612.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.9777552604675295,
|
|
"epoch": 0.06517037816353966,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.6564,
|
|
"mean_token_accuracy": 0.11448713690042496,
|
|
"num_tokens": 2397198.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.946916341781616,
|
|
"epoch": 0.06546262201180665,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.6935,
|
|
"mean_token_accuracy": 0.11213591918349267,
|
|
"num_tokens": 2407816.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.891662502288819,
|
|
"epoch": 0.06575486586007365,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.6816,
|
|
"mean_token_accuracy": 0.10039346590638161,
|
|
"num_tokens": 2419537.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.98465371131897,
|
|
"epoch": 0.06604710970834064,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.6259,
|
|
"mean_token_accuracy": 0.10811248049139977,
|
|
"num_tokens": 2429043.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.832172632217407,
|
|
"epoch": 0.06633935355660764,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.6566,
|
|
"mean_token_accuracy": 0.11272674351930619,
|
|
"num_tokens": 2439469.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.94088830947876,
|
|
"epoch": 0.06663159740487462,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.6637,
|
|
"mean_token_accuracy": 0.1104379877448082,
|
|
"num_tokens": 2449831.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.871842384338379,
|
|
"epoch": 0.06692384125314162,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.6029,
|
|
"mean_token_accuracy": 0.1185630053281784,
|
|
"num_tokens": 2460709.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.824612331390381,
|
|
"epoch": 0.06721608510140861,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.6444,
|
|
"mean_token_accuracy": 0.11218708977103234,
|
|
"num_tokens": 2472246.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.884087419509887,
|
|
"epoch": 0.06750832894967561,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.6262,
|
|
"mean_token_accuracy": 0.11107529774308204,
|
|
"num_tokens": 2482793.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.801684379577637,
|
|
"epoch": 0.0678005727979426,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.6194,
|
|
"mean_token_accuracy": 0.117616818100214,
|
|
"num_tokens": 2493186.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.965762662887573,
|
|
"epoch": 0.0680928166462096,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.6642,
|
|
"mean_token_accuracy": 0.1107054591178894,
|
|
"num_tokens": 2505100.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.8852849960327145,
|
|
"epoch": 0.0683850604944766,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.6706,
|
|
"mean_token_accuracy": 0.10744462683796882,
|
|
"num_tokens": 2515373.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.896916151046753,
|
|
"epoch": 0.06867730434274359,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.6578,
|
|
"mean_token_accuracy": 0.11280061826109886,
|
|
"num_tokens": 2526645.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.876949310302734,
|
|
"epoch": 0.06896954819101057,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.6726,
|
|
"mean_token_accuracy": 0.10913796946406365,
|
|
"num_tokens": 2537109.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.938536024093628,
|
|
"epoch": 0.06926179203927757,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 6.6137,
|
|
"mean_token_accuracy": 0.11598925217986107,
|
|
"num_tokens": 2548709.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.8740095615386965,
|
|
"epoch": 0.06955403588754457,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.6629,
|
|
"mean_token_accuracy": 0.1084805577993393,
|
|
"num_tokens": 2559802.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.837321758270264,
|
|
"epoch": 0.06984627973581156,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.6605,
|
|
"mean_token_accuracy": 0.10946222245693207,
|
|
"num_tokens": 2570343.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.8656223773956295,
|
|
"epoch": 0.07013852358407856,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.6447,
|
|
"mean_token_accuracy": 0.10875394120812416,
|
|
"num_tokens": 2580772.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.828289127349853,
|
|
"epoch": 0.07043076743234555,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.6375,
|
|
"mean_token_accuracy": 0.11186778992414474,
|
|
"num_tokens": 2592383.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.913778018951416,
|
|
"epoch": 0.07072301128061255,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.6364,
|
|
"mean_token_accuracy": 0.10910762697458268,
|
|
"num_tokens": 2602807.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.835033416748047,
|
|
"epoch": 0.07101525512887953,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.629,
|
|
"mean_token_accuracy": 0.11060894504189492,
|
|
"num_tokens": 2613978.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.906363582611084,
|
|
"epoch": 0.07130749897714653,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.6075,
|
|
"mean_token_accuracy": 0.11025396510958671,
|
|
"num_tokens": 2624044.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.882346868515015,
|
|
"epoch": 0.07159974282541352,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.6085,
|
|
"mean_token_accuracy": 0.11288385465741158,
|
|
"num_tokens": 2635550.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.821655368804931,
|
|
"epoch": 0.07189198667368052,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.5774,
|
|
"mean_token_accuracy": 0.11880745887756347,
|
|
"num_tokens": 2645343.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.8402406692504885,
|
|
"epoch": 0.07218423052194751,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.6189,
|
|
"mean_token_accuracy": 0.11299417838454247,
|
|
"num_tokens": 2655366.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.900226879119873,
|
|
"epoch": 0.07247647437021451,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.6044,
|
|
"mean_token_accuracy": 0.11406284347176551,
|
|
"num_tokens": 2666438.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.799708652496338,
|
|
"epoch": 0.0727687182184815,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.5675,
|
|
"mean_token_accuracy": 0.1110381007194519,
|
|
"num_tokens": 2677800.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.799835634231568,
|
|
"epoch": 0.07306096206674849,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 6.5575,
|
|
"mean_token_accuracy": 0.1164891928434372,
|
|
"num_tokens": 2688876.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.823706436157226,
|
|
"epoch": 0.07335320591501548,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.5231,
|
|
"mean_token_accuracy": 0.12699810415506363,
|
|
"num_tokens": 2700265.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.76237301826477,
|
|
"epoch": 0.07364544976328248,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.555,
|
|
"mean_token_accuracy": 0.12002863585948945,
|
|
"num_tokens": 2711336.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.826159906387329,
|
|
"epoch": 0.07393769361154948,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.6281,
|
|
"mean_token_accuracy": 0.11225389763712883,
|
|
"num_tokens": 2722138.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.837359285354614,
|
|
"epoch": 0.07422993745981647,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 6.5622,
|
|
"mean_token_accuracy": 0.11845961436629296,
|
|
"num_tokens": 2733016.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.825304698944092,
|
|
"epoch": 0.07452218130808347,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 6.5785,
|
|
"mean_token_accuracy": 0.11900259032845498,
|
|
"num_tokens": 2743424.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.865557479858398,
|
|
"epoch": 0.07481442515635046,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.616,
|
|
"mean_token_accuracy": 0.11359821557998658,
|
|
"num_tokens": 2753867.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.791312026977539,
|
|
"epoch": 0.07510666900461745,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.5132,
|
|
"mean_token_accuracy": 0.11576777994632721,
|
|
"num_tokens": 2764602.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.750291347503662,
|
|
"epoch": 0.07539891285288444,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 6.627,
|
|
"mean_token_accuracy": 0.11037226766347885,
|
|
"num_tokens": 2775907.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.848241567611694,
|
|
"epoch": 0.07569115670115144,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 6.5906,
|
|
"mean_token_accuracy": 0.11550377830862998,
|
|
"num_tokens": 2787455.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.756697750091552,
|
|
"epoch": 0.07598340054941843,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.5252,
|
|
"mean_token_accuracy": 0.12691028639674187,
|
|
"num_tokens": 2797600.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.748210525512695,
|
|
"epoch": 0.07627564439768543,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.4557,
|
|
"mean_token_accuracy": 0.12226552590727806,
|
|
"num_tokens": 2808056.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.778938245773316,
|
|
"epoch": 0.07656788824595243,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 6.511,
|
|
"mean_token_accuracy": 0.11900310069322587,
|
|
"num_tokens": 2817962.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.7642310619354244,
|
|
"epoch": 0.07686013209421942,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.5278,
|
|
"mean_token_accuracy": 0.12055002972483635,
|
|
"num_tokens": 2828769.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.761574554443359,
|
|
"epoch": 0.07715237594248642,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.4627,
|
|
"mean_token_accuracy": 0.12216627746820449,
|
|
"num_tokens": 2838983.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.7704860210418705,
|
|
"epoch": 0.0774446197907534,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.5237,
|
|
"mean_token_accuracy": 0.10742606818675995,
|
|
"num_tokens": 2848892.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.73878755569458,
|
|
"epoch": 0.0777368636390204,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 6.471,
|
|
"mean_token_accuracy": 0.12039833962917328,
|
|
"num_tokens": 2859303.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.816798734664917,
|
|
"epoch": 0.07802910748728739,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 6.696,
|
|
"mean_token_accuracy": 0.10775517970323563,
|
|
"num_tokens": 2871318.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.841889810562134,
|
|
"epoch": 0.07832135133555439,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 6.516,
|
|
"mean_token_accuracy": 0.11806261539459229,
|
|
"num_tokens": 2882150.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.674707317352295,
|
|
"epoch": 0.07861359518382138,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.4974,
|
|
"mean_token_accuracy": 0.1205772653222084,
|
|
"num_tokens": 2892657.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.792233276367187,
|
|
"epoch": 0.07890583903208838,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.4611,
|
|
"mean_token_accuracy": 0.12206808105111122,
|
|
"num_tokens": 2903298.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.646578741073609,
|
|
"epoch": 0.07919808288035537,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 6.5163,
|
|
"mean_token_accuracy": 0.11933484375476837,
|
|
"num_tokens": 2914684.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.808360195159912,
|
|
"epoch": 0.07949032672862236,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.4423,
|
|
"mean_token_accuracy": 0.12348946481943131,
|
|
"num_tokens": 2925191.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.701859855651856,
|
|
"epoch": 0.07978257057688935,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.5226,
|
|
"mean_token_accuracy": 0.12045117989182472,
|
|
"num_tokens": 2935498.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.793068552017212,
|
|
"epoch": 0.08007481442515635,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.4408,
|
|
"mean_token_accuracy": 0.12177050709724427,
|
|
"num_tokens": 2945590.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.6989601135253904,
|
|
"epoch": 0.08036705827342334,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 6.415,
|
|
"mean_token_accuracy": 0.1295503109693527,
|
|
"num_tokens": 2954959.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.739167213439941,
|
|
"epoch": 0.08065930212169034,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 6.4962,
|
|
"mean_token_accuracy": 0.12157377004623413,
|
|
"num_tokens": 2964808.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.74145622253418,
|
|
"epoch": 0.08095154596995734,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.4832,
|
|
"mean_token_accuracy": 0.12358191087841988,
|
|
"num_tokens": 2975032.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.724349927902222,
|
|
"epoch": 0.08124378981822433,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 6.4382,
|
|
"mean_token_accuracy": 0.12445014715194702,
|
|
"num_tokens": 2984426.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.618544483184815,
|
|
"epoch": 0.08153603366649131,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.5059,
|
|
"mean_token_accuracy": 0.11675598993897437,
|
|
"num_tokens": 2994871.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.759203100204468,
|
|
"epoch": 0.08182827751475831,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.4283,
|
|
"mean_token_accuracy": 0.1238317184150219,
|
|
"num_tokens": 3004800.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.709669494628907,
|
|
"epoch": 0.0821205213630253,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 6.5211,
|
|
"mean_token_accuracy": 0.11378786638379097,
|
|
"num_tokens": 3015561.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.706784200668335,
|
|
"epoch": 0.0824127652112923,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.4702,
|
|
"mean_token_accuracy": 0.11403566598892212,
|
|
"num_tokens": 3025885.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.691977500915527,
|
|
"epoch": 0.0827050090595593,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.4026,
|
|
"mean_token_accuracy": 0.12606138437986375,
|
|
"num_tokens": 3035524.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.703480625152588,
|
|
"epoch": 0.0829972529078263,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.5224,
|
|
"mean_token_accuracy": 0.11795256957411766,
|
|
"num_tokens": 3046957.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.78633394241333,
|
|
"epoch": 0.08328949675609329,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.4441,
|
|
"mean_token_accuracy": 0.11978953778743744,
|
|
"num_tokens": 3056713.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.641093826293945,
|
|
"epoch": 0.08358174060436027,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 6.432,
|
|
"mean_token_accuracy": 0.12080147936940193,
|
|
"num_tokens": 3068369.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.655155420303345,
|
|
"epoch": 0.08387398445262727,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.4762,
|
|
"mean_token_accuracy": 0.12348118275403977,
|
|
"num_tokens": 3079083.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.710379791259766,
|
|
"epoch": 0.08416622830089426,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.5061,
|
|
"mean_token_accuracy": 0.11596815660595894,
|
|
"num_tokens": 3089150.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.672757816314697,
|
|
"epoch": 0.08445847214916126,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 6.416,
|
|
"mean_token_accuracy": 0.12834767252206802,
|
|
"num_tokens": 3098696.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.736963558197021,
|
|
"epoch": 0.08475071599742826,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.4355,
|
|
"mean_token_accuracy": 0.120185237377882,
|
|
"num_tokens": 3108879.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.7161966323852536,
|
|
"epoch": 0.08504295984569525,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.3953,
|
|
"mean_token_accuracy": 0.12836327031254768,
|
|
"num_tokens": 3118756.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.680994129180908,
|
|
"epoch": 0.08533520369396225,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 6.4487,
|
|
"mean_token_accuracy": 0.1232797771692276,
|
|
"num_tokens": 3129130.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.678233671188354,
|
|
"epoch": 0.08562744754222923,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.4116,
|
|
"mean_token_accuracy": 0.12097690925002098,
|
|
"num_tokens": 3139528.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.644343137741089,
|
|
"epoch": 0.08591969139049623,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.4864,
|
|
"mean_token_accuracy": 0.12138932570815086,
|
|
"num_tokens": 3149456.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.659511518478394,
|
|
"epoch": 0.08621193523876322,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.4144,
|
|
"mean_token_accuracy": 0.12389760985970497,
|
|
"num_tokens": 3159851.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.634909582138062,
|
|
"epoch": 0.08650417908703022,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 6.3856,
|
|
"mean_token_accuracy": 0.13249688521027564,
|
|
"num_tokens": 3169897.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.706868124008179,
|
|
"epoch": 0.08679642293529721,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.4994,
|
|
"mean_token_accuracy": 0.122072122246027,
|
|
"num_tokens": 3180696.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.659936094284058,
|
|
"epoch": 0.08708866678356421,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 6.4162,
|
|
"mean_token_accuracy": 0.12308593988418579,
|
|
"num_tokens": 3191836.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.62534909248352,
|
|
"epoch": 0.0873809106318312,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 6.3726,
|
|
"mean_token_accuracy": 0.12648854553699493,
|
|
"num_tokens": 3202009.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.55533013343811,
|
|
"epoch": 0.0876731544800982,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 6.3333,
|
|
"mean_token_accuracy": 0.13207703083753586,
|
|
"num_tokens": 3212707.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.573125410079956,
|
|
"epoch": 0.08796539832836518,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 6.3864,
|
|
"mean_token_accuracy": 0.12348204627633094,
|
|
"num_tokens": 3223007.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.721341896057129,
|
|
"epoch": 0.08825764217663218,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004687867899220914,
|
|
"loss": 6.4061,
|
|
"mean_token_accuracy": 0.12432565912604332,
|
|
"num_tokens": 3234634.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.620856046676636,
|
|
"epoch": 0.08854988602489917,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046818549821305846,
|
|
"loss": 6.3851,
|
|
"mean_token_accuracy": 0.12274984940886498,
|
|
"num_tokens": 3245338.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.665502977371216,
|
|
"epoch": 0.08884212987316617,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046757891022600494,
|
|
"loss": 6.4692,
|
|
"mean_token_accuracy": 0.12268238142132759,
|
|
"num_tokens": 3255776.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.649141645431518,
|
|
"epoch": 0.08913437372143317,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004669670425908471,
|
|
"loss": 6.3653,
|
|
"mean_token_accuracy": 0.13077931180596353,
|
|
"num_tokens": 3265908.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.644438076019287,
|
|
"epoch": 0.08942661756970016,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004663499120822451,
|
|
"loss": 6.4581,
|
|
"mean_token_accuracy": 0.12222805321216583,
|
|
"num_tokens": 3277623.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.597890853881836,
|
|
"epoch": 0.08971886141796716,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004657275356191437,
|
|
"loss": 6.3924,
|
|
"mean_token_accuracy": 0.1266896143555641,
|
|
"num_tokens": 3288804.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.572530078887939,
|
|
"epoch": 0.09001110526623414,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00046509993026430804,
|
|
"loss": 6.4006,
|
|
"mean_token_accuracy": 0.1275424525141716,
|
|
"num_tokens": 3299162.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.724449968338012,
|
|
"epoch": 0.09030334911450114,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004644671132238558,
|
|
"loss": 6.4128,
|
|
"mean_token_accuracy": 0.12625961378216743,
|
|
"num_tokens": 3311425.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.537902307510376,
|
|
"epoch": 0.09059559296276813,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00046382910184678585,
|
|
"loss": 6.3823,
|
|
"mean_token_accuracy": 0.12513037025928497,
|
|
"num_tokens": 3322623.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.771864175796509,
|
|
"epoch": 0.09088783681103513,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004631859136245025,
|
|
"loss": 6.3973,
|
|
"mean_token_accuracy": 0.12243670225143433,
|
|
"num_tokens": 3333234.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.626861143112182,
|
|
"epoch": 0.09118008065930212,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004625375661903357,
|
|
"loss": 6.3902,
|
|
"mean_token_accuracy": 0.12339795604348183,
|
|
"num_tokens": 3343302.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.566328763961792,
|
|
"epoch": 0.09147232450756912,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 6.3709,
|
|
"mean_token_accuracy": 0.11957384571433068,
|
|
"num_tokens": 3355107.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.633986139297486,
|
|
"epoch": 0.09176456835583612,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046122546492639643,
|
|
"loss": 6.3414,
|
|
"mean_token_accuracy": 0.13049424216151237,
|
|
"num_tokens": 3366459.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.5118812084197994,
|
|
"epoch": 0.0920568122041031,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000460561747068543,
|
|
"loss": 6.319,
|
|
"mean_token_accuracy": 0.13351040184497834,
|
|
"num_tokens": 3377050.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.558924579620362,
|
|
"epoch": 0.0923490560523701,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004598929419416578,
|
|
"loss": 6.3246,
|
|
"mean_token_accuracy": 0.1303851567208767,
|
|
"num_tokens": 3387810.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.566503572463989,
|
|
"epoch": 0.09264129990063709,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045921906788137123,
|
|
"loss": 6.2929,
|
|
"mean_token_accuracy": 0.13062896132469176,
|
|
"num_tokens": 3398038.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.659072780609131,
|
|
"epoch": 0.09293354374890408,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045854014336228115,
|
|
"loss": 6.3714,
|
|
"mean_token_accuracy": 0.1278992585837841,
|
|
"num_tokens": 3408426.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.563223266601563,
|
|
"epoch": 0.09322578759717108,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045785618699744615,
|
|
"loss": 6.3084,
|
|
"mean_token_accuracy": 0.13401768282055854,
|
|
"num_tokens": 3419268.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.621515703201294,
|
|
"epoch": 0.09351803144543808,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00045716721753787543,
|
|
"loss": 6.3067,
|
|
"mean_token_accuracy": 0.13638126030564307,
|
|
"num_tokens": 3429813.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.482415342330933,
|
|
"epoch": 0.09381027529370507,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004564732538720148,
|
|
"loss": 6.2883,
|
|
"mean_token_accuracy": 0.13472536355257034,
|
|
"num_tokens": 3440062.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.672563982009888,
|
|
"epoch": 0.09410251914197205,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00045577431502522877,
|
|
"loss": 6.2849,
|
|
"mean_token_accuracy": 0.1270398922264576,
|
|
"num_tokens": 3450204.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.488152503967285,
|
|
"epoch": 0.09439476299023905,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004550704201592787,
|
|
"loss": 6.3027,
|
|
"mean_token_accuracy": 0.12851924523711206,
|
|
"num_tokens": 3460593.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.628467226028443,
|
|
"epoch": 0.09468700683850605,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004543615885717981,
|
|
"loss": 6.3229,
|
|
"mean_token_accuracy": 0.1320893421769142,
|
|
"num_tokens": 3471617.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.452277565002442,
|
|
"epoch": 0.09497925068677304,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 6.3045,
|
|
"mean_token_accuracy": 0.12825360149145126,
|
|
"num_tokens": 3481701.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.581420993804931,
|
|
"epoch": 0.09527149453504004,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004529291930989592,
|
|
"loss": 6.3309,
|
|
"mean_token_accuracy": 0.12806977778673173,
|
|
"num_tokens": 3491148.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.6200621128082275,
|
|
"epoch": 0.09556373838330703,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004522056684834464,
|
|
"loss": 6.3644,
|
|
"mean_token_accuracy": 0.12346012219786644,
|
|
"num_tokens": 3501675.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.554399871826172,
|
|
"epoch": 0.09585598223157403,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004514772856850173,
|
|
"loss": 6.3077,
|
|
"mean_token_accuracy": 0.13186056092381476,
|
|
"num_tokens": 3512988.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.5387852668762205,
|
|
"epoch": 0.09614822607984103,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004507440646726542,
|
|
"loss": 6.3082,
|
|
"mean_token_accuracy": 0.12828391939401626,
|
|
"num_tokens": 3524004.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.525148534774781,
|
|
"epoch": 0.09644046992810801,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004500060255479818,
|
|
"loss": 6.313,
|
|
"mean_token_accuracy": 0.12847139537334443,
|
|
"num_tokens": 3533627.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.571453475952149,
|
|
"epoch": 0.096732713776375,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004492631885447151,
|
|
"loss": 6.301,
|
|
"mean_token_accuracy": 0.127740541100502,
|
|
"num_tokens": 3544798.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.584779834747314,
|
|
"epoch": 0.097024957624642,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00044851557402810616,
|
|
"loss": 6.404,
|
|
"mean_token_accuracy": 0.12892045825719833,
|
|
"num_tokens": 3556379.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.601266098022461,
|
|
"epoch": 0.097317201472909,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044776320249438444,
|
|
"loss": 6.2909,
|
|
"mean_token_accuracy": 0.1332499749958515,
|
|
"num_tokens": 3567032.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.583350610733032,
|
|
"epoch": 0.09760944532117599,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00044700609457019565,
|
|
"loss": 6.3648,
|
|
"mean_token_accuracy": 0.12196323797106742,
|
|
"num_tokens": 3578419.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.524851226806641,
|
|
"epoch": 0.09790168916944299,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004462442710120359,
|
|
"loss": 6.2726,
|
|
"mean_token_accuracy": 0.13355116918683052,
|
|
"num_tokens": 3589501.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.51496729850769,
|
|
"epoch": 0.09819393301770998,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000445477752705683,
|
|
"loss": 6.2603,
|
|
"mean_token_accuracy": 0.13476647436618805,
|
|
"num_tokens": 3600162.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.532532167434693,
|
|
"epoch": 0.09848617686597697,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 6.3614,
|
|
"mean_token_accuracy": 0.1309310808777809,
|
|
"num_tokens": 3610860.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.55736231803894,
|
|
"epoch": 0.09877842071424396,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004439307160344765,
|
|
"loss": 6.3372,
|
|
"mean_token_accuracy": 0.1256033793091774,
|
|
"num_tokens": 3620989.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.520544958114624,
|
|
"epoch": 0.09907066456251096,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00044315024008241473,
|
|
"loss": 6.2777,
|
|
"mean_token_accuracy": 0.1307544231414795,
|
|
"num_tokens": 3631602.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.575679063796997,
|
|
"epoch": 0.09936290841077795,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004423651542065806,
|
|
"loss": 6.3269,
|
|
"mean_token_accuracy": 0.1257415808737278,
|
|
"num_tokens": 3642326.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.539660263061523,
|
|
"epoch": 0.09965515225904495,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044157547993050006,
|
|
"loss": 6.2822,
|
|
"mean_token_accuracy": 0.1356342352926731,
|
|
"num_tokens": 3652036.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.561394262313843,
|
|
"epoch": 0.09994739610731194,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044078123890349227,
|
|
"loss": 6.3283,
|
|
"mean_token_accuracy": 0.13123508021235467,
|
|
"num_tokens": 3661557.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.59216947555542,
|
|
"epoch": 0.10023963995557894,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00043998245290007606,
|
|
"loss": 6.3135,
|
|
"mean_token_accuracy": 0.13179217875003815,
|
|
"num_tokens": 3671487.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.4940907001495365,
|
|
"epoch": 0.10053188380384592,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00043917914381937323,
|
|
"loss": 6.3445,
|
|
"mean_token_accuracy": 0.1277608558535576,
|
|
"num_tokens": 3682873.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.547703456878662,
|
|
"epoch": 0.10082412765211292,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00043837133368450815,
|
|
"loss": 6.2517,
|
|
"mean_token_accuracy": 0.1312596969306469,
|
|
"num_tokens": 3693635.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.48290376663208,
|
|
"epoch": 0.10111637150037991,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004375590446420037,
|
|
"loss": 6.2628,
|
|
"mean_token_accuracy": 0.1380753017961979,
|
|
"num_tokens": 3703777.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.5642343044281,
|
|
"epoch": 0.10140861534864691,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004367422989611743,
|
|
"loss": 6.2828,
|
|
"mean_token_accuracy": 0.13648706600069999,
|
|
"num_tokens": 3716370.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.527584552764893,
|
|
"epoch": 0.1017008591969139,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004359211190335153,
|
|
"loss": 6.3367,
|
|
"mean_token_accuracy": 0.13259412199258805,
|
|
"num_tokens": 3726473.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.588420152664185,
|
|
"epoch": 0.1019931030451809,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 6.3462,
|
|
"mean_token_accuracy": 0.12390820607542992,
|
|
"num_tokens": 3737645.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.521403980255127,
|
|
"epoch": 0.1022853468934479,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00043426554661090853,
|
|
"loss": 6.3338,
|
|
"mean_token_accuracy": 0.13016941174864768,
|
|
"num_tokens": 3749347.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.540117597579956,
|
|
"epoch": 0.10257759074171488,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043343119950431516,
|
|
"loss": 6.3035,
|
|
"mean_token_accuracy": 0.13260690197348596,
|
|
"num_tokens": 3760131.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.532932329177856,
|
|
"epoch": 0.10286983458998188,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043259250892635644,
|
|
"loss": 6.2826,
|
|
"mean_token_accuracy": 0.12948739156126976,
|
|
"num_tokens": 3770722.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.5060454368591305,
|
|
"epoch": 0.10316207843824887,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004317494978701582,
|
|
"loss": 6.2532,
|
|
"mean_token_accuracy": 0.14073092937469484,
|
|
"num_tokens": 3781562.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.506136131286621,
|
|
"epoch": 0.10345432228651587,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004309021894472943,
|
|
"loss": 6.2594,
|
|
"mean_token_accuracy": 0.13220234960317612,
|
|
"num_tokens": 3791029.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.534422826766968,
|
|
"epoch": 0.10374656613478286,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004300506068871534,
|
|
"loss": 6.2559,
|
|
"mean_token_accuracy": 0.13380615487694741,
|
|
"num_tokens": 3802526.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.4779705047607425,
|
|
"epoch": 0.10403880998304986,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042919477353630135,
|
|
"loss": 6.2275,
|
|
"mean_token_accuracy": 0.13506562113761902,
|
|
"num_tokens": 3813394.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.496424818038941,
|
|
"epoch": 0.10433105383131686,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000428334712857842,
|
|
"loss": 6.2403,
|
|
"mean_token_accuracy": 0.1317013218998909,
|
|
"num_tokens": 3824852.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.487008571624756,
|
|
"epoch": 0.10462329767958385,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00042747044843077304,
|
|
"loss": 6.2393,
|
|
"mean_token_accuracy": 0.13541778847575187,
|
|
"num_tokens": 3835743.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.498304033279419,
|
|
"epoch": 0.10491554152785083,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00042660200394934047,
|
|
"loss": 6.2294,
|
|
"mean_token_accuracy": 0.13833673521876336,
|
|
"num_tokens": 3846897.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.46450982093811,
|
|
"epoch": 0.10520778537611783,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00042572940322238844,
|
|
"loss": 6.2409,
|
|
"mean_token_accuracy": 0.13357219472527504,
|
|
"num_tokens": 3857351.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.4254958152771,
|
|
"epoch": 0.10550002922438483,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 6.2055,
|
|
"mean_token_accuracy": 0.13916457816958427,
|
|
"num_tokens": 3867696.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.512742614746093,
|
|
"epoch": 0.10579227307265182,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004239718288363745,
|
|
"loss": 6.3049,
|
|
"mean_token_accuracy": 0.13644790649414062,
|
|
"num_tokens": 3878473.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.476879119873047,
|
|
"epoch": 0.10608451692091882,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004230869033621023,
|
|
"loss": 6.2803,
|
|
"mean_token_accuracy": 0.1361045390367508,
|
|
"num_tokens": 3888750.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.549038934707641,
|
|
"epoch": 0.10637676076918581,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004221979180105688,
|
|
"loss": 6.2918,
|
|
"mean_token_accuracy": 0.13268638402223587,
|
|
"num_tokens": 3898867.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.439270830154419,
|
|
"epoch": 0.10666900461745281,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042130489715375645,
|
|
"loss": 6.2231,
|
|
"mean_token_accuracy": 0.13936343863606454,
|
|
"num_tokens": 3909103.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.49756350517273,
|
|
"epoch": 0.10696124846571979,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00042040786527428335,
|
|
"loss": 6.1534,
|
|
"mean_token_accuracy": 0.13987711891531945,
|
|
"num_tokens": 3919592.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.52377724647522,
|
|
"epoch": 0.10725349231398679,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004195068469647315,
|
|
"loss": 6.2452,
|
|
"mean_token_accuracy": 0.13477420955896377,
|
|
"num_tokens": 3930730.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.382622146606446,
|
|
"epoch": 0.10754573616225378,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00041860186692697297,
|
|
"loss": 6.2535,
|
|
"mean_token_accuracy": 0.130735794454813,
|
|
"num_tokens": 3942535.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.54898157119751,
|
|
"epoch": 0.10783798001052078,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00041769294997149264,
|
|
"loss": 6.2364,
|
|
"mean_token_accuracy": 0.1324638806283474,
|
|
"num_tokens": 3953651.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.497935390472412,
|
|
"epoch": 0.10813022385878777,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004167801210167081,
|
|
"loss": 6.2087,
|
|
"mean_token_accuracy": 0.13810430020093917,
|
|
"num_tokens": 3963501.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.502410459518432,
|
|
"epoch": 0.10842246770705477,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004158634050882861,
|
|
"loss": 6.2066,
|
|
"mean_token_accuracy": 0.1336246259510517,
|
|
"num_tokens": 3973617.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.448571395874024,
|
|
"epoch": 0.10871471155532177,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004149428273184569,
|
|
"loss": 6.2734,
|
|
"mean_token_accuracy": 0.12727659419178963,
|
|
"num_tokens": 3985313.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.5132608890533445,
|
|
"epoch": 0.10900695540358875,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 6.2132,
|
|
"mean_token_accuracy": 0.1299443393945694,
|
|
"num_tokens": 3995136.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.359571886062622,
|
|
"epoch": 0.10929919925185574,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000413090187312178,
|
|
"loss": 6.1079,
|
|
"mean_token_accuracy": 0.15453708469867705,
|
|
"num_tokens": 4004633.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.430087947845459,
|
|
"epoch": 0.10959144310012274,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004121581758667898,
|
|
"loss": 6.215,
|
|
"mean_token_accuracy": 0.13473245427012442,
|
|
"num_tokens": 4013949.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.497232055664062,
|
|
"epoch": 0.10988368694838974,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00041122240416072533,
|
|
"loss": 6.1965,
|
|
"mean_token_accuracy": 0.13759151250123977,
|
|
"num_tokens": 4024456.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.374862003326416,
|
|
"epoch": 0.11017593079665673,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004102828978486385,
|
|
"loss": 6.1864,
|
|
"mean_token_accuracy": 0.14066949412226676,
|
|
"num_tokens": 4035271.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.514592981338501,
|
|
"epoch": 0.11046817464492373,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004093396826875695,
|
|
"loss": 6.2668,
|
|
"mean_token_accuracy": 0.1375574715435505,
|
|
"num_tokens": 4045861.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.489775943756103,
|
|
"epoch": 0.11076041849319072,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00040839278453623837,
|
|
"loss": 6.1744,
|
|
"mean_token_accuracy": 0.13124485611915587,
|
|
"num_tokens": 4056060.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.384185409545898,
|
|
"epoch": 0.1110526623414577,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004074422293543363,
|
|
"loss": 6.2402,
|
|
"mean_token_accuracy": 0.13457444533705712,
|
|
"num_tokens": 4068098.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.451984453201294,
|
|
"epoch": 0.1113449061897247,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004064880432018137,
|
|
"loss": 6.154,
|
|
"mean_token_accuracy": 0.15216126739978791,
|
|
"num_tokens": 4079186.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.405668115615844,
|
|
"epoch": 0.1116371500379917,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00040553025223816615,
|
|
"loss": 6.1575,
|
|
"mean_token_accuracy": 0.14098218828439713,
|
|
"num_tokens": 4089430.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.461432552337646,
|
|
"epoch": 0.1119293938862587,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00040456888272171653,
|
|
"loss": 6.1631,
|
|
"mean_token_accuracy": 0.14054136276245116,
|
|
"num_tokens": 4099702.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.486164999008179,
|
|
"epoch": 0.11222163773452569,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00040360396100889577,
|
|
"loss": 6.2735,
|
|
"mean_token_accuracy": 0.127665276825428,
|
|
"num_tokens": 4111108.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.47590913772583,
|
|
"epoch": 0.11251388158279269,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 6.2286,
|
|
"mean_token_accuracy": 0.13972759246826172,
|
|
"num_tokens": 4122470.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.409267807006836,
|
|
"epoch": 0.11280612543105968,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000401663566906066,
|
|
"loss": 6.1722,
|
|
"mean_token_accuracy": 0.13970772325992584,
|
|
"num_tokens": 4132652.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.41186146736145,
|
|
"epoch": 0.11309836927932668,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00040068814771294134,
|
|
"loss": 6.253,
|
|
"mean_token_accuracy": 0.1293734610080719,
|
|
"num_tokens": 4143449.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.475877523422241,
|
|
"epoch": 0.11339061312759366,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003997092827157562,
|
|
"loss": 6.1805,
|
|
"mean_token_accuracy": 0.14184334874153137,
|
|
"num_tokens": 4153536.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.463614416122437,
|
|
"epoch": 0.11368285697586065,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000398726998750589,
|
|
"loss": 6.2478,
|
|
"mean_token_accuracy": 0.1280618853867054,
|
|
"num_tokens": 4165339.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.478766345977784,
|
|
"epoch": 0.11397510082412765,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00039774132274725076,
|
|
"loss": 6.2953,
|
|
"mean_token_accuracy": 0.13131710588932038,
|
|
"num_tokens": 4177171.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.531812000274658,
|
|
"epoch": 0.11426734467239465,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039675228172854707,
|
|
"loss": 6.1826,
|
|
"mean_token_accuracy": 0.14128784090280533,
|
|
"num_tokens": 4186577.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.3304266929626465,
|
|
"epoch": 0.11455958852066164,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003957599028095371,
|
|
"loss": 6.2077,
|
|
"mean_token_accuracy": 0.13591980412602425,
|
|
"num_tokens": 4196962.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.440403938293457,
|
|
"epoch": 0.11485183236892864,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00039476421319679017,
|
|
"loss": 6.202,
|
|
"mean_token_accuracy": 0.139873356372118,
|
|
"num_tokens": 4207544.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.491571664810181,
|
|
"epoch": 0.11514407621719563,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00039376524018764,
|
|
"loss": 6.1867,
|
|
"mean_token_accuracy": 0.138301333039999,
|
|
"num_tokens": 4217359.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.42297248840332,
|
|
"epoch": 0.11543632006546262,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00039276301116943616,
|
|
"loss": 6.1981,
|
|
"mean_token_accuracy": 0.13637675866484641,
|
|
"num_tokens": 4227826.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.479195499420166,
|
|
"epoch": 0.11572856391372961,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003917575536187936,
|
|
"loss": 6.1559,
|
|
"mean_token_accuracy": 0.13742526918649672,
|
|
"num_tokens": 4238591.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.44315733909607,
|
|
"epoch": 0.11602080776199661,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 6.1963,
|
|
"mean_token_accuracy": 0.13515803813934327,
|
|
"num_tokens": 4248771.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.484133291244507,
|
|
"epoch": 0.1163130516102636,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00038973706326845495,
|
|
"loss": 6.254,
|
|
"mean_token_accuracy": 0.13710065484046935,
|
|
"num_tokens": 4259234.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.458966588973999,
|
|
"epoch": 0.1166052954585306,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0003887220858615225,
|
|
"loss": 6.2597,
|
|
"mean_token_accuracy": 0.1325739674270153,
|
|
"num_tokens": 4270891.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.502737188339234,
|
|
"epoch": 0.1168975393067976,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003877039907061597,
|
|
"loss": 6.271,
|
|
"mean_token_accuracy": 0.1323829747736454,
|
|
"num_tokens": 4282893.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.417344093322754,
|
|
"epoch": 0.11718978315506459,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003866828057139598,
|
|
"loss": 6.1572,
|
|
"mean_token_accuracy": 0.13396240249276162,
|
|
"num_tokens": 4292630.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.377220344543457,
|
|
"epoch": 0.11748202700333157,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00038565855888122503,
|
|
"loss": 6.1991,
|
|
"mean_token_accuracy": 0.13958619087934493,
|
|
"num_tokens": 4302849.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.4756178855896,
|
|
"epoch": 0.11777427085159857,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00038463127828819975,
|
|
"loss": 6.2179,
|
|
"mean_token_accuracy": 0.13426078259944915,
|
|
"num_tokens": 4313178.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.52903003692627,
|
|
"epoch": 0.11806651469986557,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00038360099209830043,
|
|
"loss": 6.2464,
|
|
"mean_token_accuracy": 0.13791208043694497,
|
|
"num_tokens": 4323205.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.386642026901245,
|
|
"epoch": 0.11835875854813256,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003825677285573433,
|
|
"loss": 6.1849,
|
|
"mean_token_accuracy": 0.1381428360939026,
|
|
"num_tokens": 4333840.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.479530143737793,
|
|
"epoch": 0.11865100239639956,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00038153151599277027,
|
|
"loss": 6.1773,
|
|
"mean_token_accuracy": 0.14190752506256105,
|
|
"num_tokens": 4344489.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.454029750823975,
|
|
"epoch": 0.11894324624466655,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003804923828128723,
|
|
"loss": 6.1564,
|
|
"mean_token_accuracy": 0.13591767027974128,
|
|
"num_tokens": 4355008.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.409417295455933,
|
|
"epoch": 0.11923549009293355,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003794503575060104,
|
|
"loss": 6.1939,
|
|
"mean_token_accuracy": 0.14090473279356958,
|
|
"num_tokens": 4365944.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.4883616924285885,
|
|
"epoch": 0.11952773394120053,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00037840546863983484,
|
|
"loss": 6.1953,
|
|
"mean_token_accuracy": 0.13460226878523826,
|
|
"num_tokens": 4376448.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.437252426147461,
|
|
"epoch": 0.11981997778946753,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003773577448605015,
|
|
"loss": 6.131,
|
|
"mean_token_accuracy": 0.14052277579903602,
|
|
"num_tokens": 4386748.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.414979887008667,
|
|
"epoch": 0.12011222163773452,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003763072148918872,
|
|
"loss": 6.1115,
|
|
"mean_token_accuracy": 0.14440772235393523,
|
|
"num_tokens": 4396956.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.390393924713135,
|
|
"epoch": 0.12040446548600152,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003752539075348017,
|
|
"loss": 6.1607,
|
|
"mean_token_accuracy": 0.1384737014770508,
|
|
"num_tokens": 4406551.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.506589317321778,
|
|
"epoch": 0.12069670933426851,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00037419785166619817,
|
|
"loss": 6.1342,
|
|
"mean_token_accuracy": 0.13821437284350396,
|
|
"num_tokens": 4416847.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.371277284622193,
|
|
"epoch": 0.12098895318253551,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003731390762383818,
|
|
"loss": 6.1307,
|
|
"mean_token_accuracy": 0.14046358093619346,
|
|
"num_tokens": 4427579.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.357118844985962,
|
|
"epoch": 0.1212811970308025,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003720776102782158,
|
|
"loss": 6.1174,
|
|
"mean_token_accuracy": 0.14415650218725204,
|
|
"num_tokens": 4438473.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.391526699066162,
|
|
"epoch": 0.12157344087906949,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00037101348288632555,
|
|
"loss": 6.1079,
|
|
"mean_token_accuracy": 0.1449679099023342,
|
|
"num_tokens": 4448851.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.311391687393188,
|
|
"epoch": 0.12186568472733648,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003699467232363012,
|
|
"loss": 6.1264,
|
|
"mean_token_accuracy": 0.14008628129959105,
|
|
"num_tokens": 4459612.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.388618898391724,
|
|
"epoch": 0.12215792857560348,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003688773605738973,
|
|
"loss": 6.107,
|
|
"mean_token_accuracy": 0.14008531495928764,
|
|
"num_tokens": 4470365.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.348719549179077,
|
|
"epoch": 0.12245017242387048,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00036780542421623134,
|
|
"loss": 6.1548,
|
|
"mean_token_accuracy": 0.14530792385339736,
|
|
"num_tokens": 4481329.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.451541805267334,
|
|
"epoch": 0.12274241627213747,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003667309435509802,
|
|
"loss": 6.1099,
|
|
"mean_token_accuracy": 0.1375292494893074,
|
|
"num_tokens": 4491904.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.369375848770142,
|
|
"epoch": 0.12303466012040447,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003656539480355741,
|
|
"loss": 6.0846,
|
|
"mean_token_accuracy": 0.14959308356046677,
|
|
"num_tokens": 4501432.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.393247985839844,
|
|
"epoch": 0.12332690396867146,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003645744671963891,
|
|
"loss": 6.0733,
|
|
"mean_token_accuracy": 0.14439773932099342,
|
|
"num_tokens": 4511512.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.381515789031982,
|
|
"epoch": 0.12361914781693846,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003634925306279376,
|
|
"loss": 6.177,
|
|
"mean_token_accuracy": 0.14277450367808342,
|
|
"num_tokens": 4521701.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.400973653793335,
|
|
"epoch": 0.12391139166520544,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003624081679920574,
|
|
"loss": 6.1464,
|
|
"mean_token_accuracy": 0.1424795500934124,
|
|
"num_tokens": 4531512.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.378998327255249,
|
|
"epoch": 0.12420363551347244,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003613214090170977,
|
|
"loss": 6.1222,
|
|
"mean_token_accuracy": 0.1437835432589054,
|
|
"num_tokens": 4542083.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.3635622501373295,
|
|
"epoch": 0.12449587936173943,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003602322834971048,
|
|
"loss": 6.1267,
|
|
"mean_token_accuracy": 0.1413314886391163,
|
|
"num_tokens": 4553207.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.4260810852050785,
|
|
"epoch": 0.12478812321000643,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0003591408212910051,
|
|
"loss": 6.1772,
|
|
"mean_token_accuracy": 0.13948365449905395,
|
|
"num_tokens": 4564349.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.413608407974243,
|
|
"epoch": 0.1250803670582734,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003580470523217863,
|
|
"loss": 6.0712,
|
|
"mean_token_accuracy": 0.14198292568325996,
|
|
"num_tokens": 4574935.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.339417600631714,
|
|
"epoch": 0.12537261090654042,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003569510065756771,
|
|
"loss": 6.1782,
|
|
"mean_token_accuracy": 0.13700612559914588,
|
|
"num_tokens": 4586439.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.337814044952393,
|
|
"epoch": 0.1256648547548074,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003558527141013254,
|
|
"loss": 6.0303,
|
|
"mean_token_accuracy": 0.151566082239151,
|
|
"num_tokens": 4596439.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.288027906417847,
|
|
"epoch": 0.1259570986030744,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003547522050089742,
|
|
"loss": 6.1032,
|
|
"mean_token_accuracy": 0.14451807215809823,
|
|
"num_tokens": 4606920.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.478337049484253,
|
|
"epoch": 0.1262493424513414,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00035364950946963606,
|
|
"loss": 6.1458,
|
|
"mean_token_accuracy": 0.14522892534732817,
|
|
"num_tokens": 4618234.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.32128791809082,
|
|
"epoch": 0.1265415862996084,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003525446577142663,
|
|
"loss": 5.9731,
|
|
"mean_token_accuracy": 0.15805664658546448,
|
|
"num_tokens": 4629201.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.391382646560669,
|
|
"epoch": 0.1268338301478754,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00035143768003293395,
|
|
"loss": 6.2195,
|
|
"mean_token_accuracy": 0.1382079988718033,
|
|
"num_tokens": 4641008.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.356924915313721,
|
|
"epoch": 0.12712607399614237,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003503286067739913,
|
|
"loss": 6.1099,
|
|
"mean_token_accuracy": 0.14097246378660203,
|
|
"num_tokens": 4651969.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.472826147079468,
|
|
"epoch": 0.12741831784440938,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00034921746834324193,
|
|
"loss": 6.1656,
|
|
"mean_token_accuracy": 0.13588185235857964,
|
|
"num_tokens": 4663081.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.306978511810303,
|
|
"epoch": 0.12771056169267636,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003481042952031072,
|
|
"loss": 5.9404,
|
|
"mean_token_accuracy": 0.15880992263555527,
|
|
"num_tokens": 4673147.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.293196773529052,
|
|
"epoch": 0.12800280554094337,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003469891178717911,
|
|
"loss": 6.0981,
|
|
"mean_token_accuracy": 0.14433757364749908,
|
|
"num_tokens": 4683237.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.412969493865967,
|
|
"epoch": 0.12829504938921035,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003458719669224436,
|
|
"loss": 6.0681,
|
|
"mean_token_accuracy": 0.14517636150121688,
|
|
"num_tokens": 4694541.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.4162555694580075,
|
|
"epoch": 0.12858729323747736,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003447528729823221,
|
|
"loss": 6.1,
|
|
"mean_token_accuracy": 0.14119071885943413,
|
|
"num_tokens": 4704583.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.2811743259429935,
|
|
"epoch": 0.12887953708574434,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003436318667319525,
|
|
"loss": 6.0803,
|
|
"mean_token_accuracy": 0.14858463555574417,
|
|
"num_tokens": 4715880.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.407436227798462,
|
|
"epoch": 0.12917178093401133,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00034250897890428716,
|
|
"loss": 6.0807,
|
|
"mean_token_accuracy": 0.14115795344114304,
|
|
"num_tokens": 4726428.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.3981640338897705,
|
|
"epoch": 0.12946402478227834,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003413842402838633,
|
|
"loss": 6.0702,
|
|
"mean_token_accuracy": 0.14687135368585585,
|
|
"num_tokens": 4736782.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.329130554199219,
|
|
"epoch": 0.12975626863054532,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00034025768170595834,
|
|
"loss": 6.1513,
|
|
"mean_token_accuracy": 0.14326717853546142,
|
|
"num_tokens": 4747084.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.4172278881073,
|
|
"epoch": 0.13004851247881233,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003391293340557446,
|
|
"loss": 6.1026,
|
|
"mean_token_accuracy": 0.13796553611755372,
|
|
"num_tokens": 4758499.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.3779843807220455,
|
|
"epoch": 0.1303407563270793,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003379992282674431,
|
|
"loss": 6.0984,
|
|
"mean_token_accuracy": 0.14021950736641883,
|
|
"num_tokens": 4768553.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.3388991355896,
|
|
"epoch": 0.13063300017534632,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003368673953234749,
|
|
"loss": 6.0441,
|
|
"mean_token_accuracy": 0.14745086207985877,
|
|
"num_tokens": 4778214.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.334152126312256,
|
|
"epoch": 0.1309252440236133,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00033573386625361176,
|
|
"loss": 6.1084,
|
|
"mean_token_accuracy": 0.13899068757891656,
|
|
"num_tokens": 4789870.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.202721500396729,
|
|
"epoch": 0.13121748787188028,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00033459867213412567,
|
|
"loss": 5.8645,
|
|
"mean_token_accuracy": 0.16954285800457,
|
|
"num_tokens": 4801320.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.27685284614563,
|
|
"epoch": 0.1315097317201473,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000333461844086937,
|
|
"loss": 6.1168,
|
|
"mean_token_accuracy": 0.14532479792833328,
|
|
"num_tokens": 4812170.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.427320957183838,
|
|
"epoch": 0.13180197556841428,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033232341327876097,
|
|
"loss": 6.1355,
|
|
"mean_token_accuracy": 0.1392444886267185,
|
|
"num_tokens": 4823279.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.399251985549927,
|
|
"epoch": 0.13209421941668129,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003311834109202531,
|
|
"loss": 5.9841,
|
|
"mean_token_accuracy": 0.15361466854810715,
|
|
"num_tokens": 4833590.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.250917291641235,
|
|
"epoch": 0.13238646326494827,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00033004186826515416,
|
|
"loss": 6.0554,
|
|
"mean_token_accuracy": 0.1447410002350807,
|
|
"num_tokens": 4844364.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.405690717697143,
|
|
"epoch": 0.13267870711321528,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003288988166094324,
|
|
"loss": 6.0467,
|
|
"mean_token_accuracy": 0.14364370033144952,
|
|
"num_tokens": 4854375.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.3763069152832035,
|
|
"epoch": 0.13297095096148226,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00032775428729042656,
|
|
"loss": 6.0646,
|
|
"mean_token_accuracy": 0.14396966099739075,
|
|
"num_tokens": 4864704.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.3314014911651615,
|
|
"epoch": 0.13326319480974924,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000326608311685986,
|
|
"loss": 6.0507,
|
|
"mean_token_accuracy": 0.15053208023309708,
|
|
"num_tokens": 4874774.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.303547191619873,
|
|
"epoch": 0.13355543865801625,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003254609212136108,
|
|
"loss": 6.0658,
|
|
"mean_token_accuracy": 0.14474221244454383,
|
|
"num_tokens": 4886263.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.270662403106689,
|
|
"epoch": 0.13384768250628323,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00032431214732959036,
|
|
"loss": 6.1063,
|
|
"mean_token_accuracy": 0.14200906977057456,
|
|
"num_tokens": 4897416.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.426070737838745,
|
|
"epoch": 0.13413992635455024,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000323162021528141,
|
|
"loss": 6.0629,
|
|
"mean_token_accuracy": 0.14902425408363343,
|
|
"num_tokens": 4907710.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.361135005950928,
|
|
"epoch": 0.13443217020281722,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00032201057534054264,
|
|
"loss": 6.1144,
|
|
"mean_token_accuracy": 0.1474320709705353,
|
|
"num_tokens": 4919190.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.297651672363282,
|
|
"epoch": 0.13472441405108423,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00032085784033427414,
|
|
"loss": 6.0906,
|
|
"mean_token_accuracy": 0.14951519295573235,
|
|
"num_tokens": 4929892.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.35400619506836,
|
|
"epoch": 0.13501665789935122,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003197038481121478,
|
|
"loss": 6.1102,
|
|
"mean_token_accuracy": 0.14383033886551858,
|
|
"num_tokens": 4940649.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.300059413909912,
|
|
"epoch": 0.1353089017476182,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003185486303114436,
|
|
"loss": 6.1523,
|
|
"mean_token_accuracy": 0.1469470590353012,
|
|
"num_tokens": 4950952.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.400362157821656,
|
|
"epoch": 0.1356011455958852,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003173922186030409,
|
|
"loss": 6.0551,
|
|
"mean_token_accuracy": 0.14830568730831145,
|
|
"num_tokens": 4961473.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.3428558826446535,
|
|
"epoch": 0.1358933894441522,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000316234644690551,
|
|
"loss": 6.1208,
|
|
"mean_token_accuracy": 0.14066094309091567,
|
|
"num_tokens": 4972702.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.3625664710998535,
|
|
"epoch": 0.1361856332924192,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0003150759403094473,
|
|
"loss": 6.1168,
|
|
"mean_token_accuracy": 0.13751328289508818,
|
|
"num_tokens": 4983003.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.311033821105957,
|
|
"epoch": 0.13647787714068618,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00031391613722619587,
|
|
"loss": 6.0712,
|
|
"mean_token_accuracy": 0.14670585095882416,
|
|
"num_tokens": 4993378.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.347900104522705,
|
|
"epoch": 0.1367701209889532,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000312755267237384,
|
|
"loss": 6.1411,
|
|
"mean_token_accuracy": 0.13584877029061318,
|
|
"num_tokens": 5004076.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.4429466247558596,
|
|
"epoch": 0.13706236483722017,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003115933621688488,
|
|
"loss": 6.1122,
|
|
"mean_token_accuracy": 0.14021623879671097,
|
|
"num_tokens": 5014718.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.398041105270385,
|
|
"epoch": 0.13735460868548718,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00031043045387480487,
|
|
"loss": 6.0959,
|
|
"mean_token_accuracy": 0.14150511622428893,
|
|
"num_tokens": 5026102.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.31099681854248,
|
|
"epoch": 0.13764685253375417,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003092665742369703,
|
|
"loss": 6.0418,
|
|
"mean_token_accuracy": 0.14680029153823854,
|
|
"num_tokens": 5036847.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.337398195266724,
|
|
"epoch": 0.13793909638202115,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00030810175516369343,
|
|
"loss": 5.9778,
|
|
"mean_token_accuracy": 0.15376632735133172,
|
|
"num_tokens": 5047751.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.33739070892334,
|
|
"epoch": 0.13823134023028816,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003069360285890775,
|
|
"loss": 6.0329,
|
|
"mean_token_accuracy": 0.144328011572361,
|
|
"num_tokens": 5057914.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.319474363327027,
|
|
"epoch": 0.13852358407855514,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00030576942647210547,
|
|
"loss": 6.0952,
|
|
"mean_token_accuracy": 0.14145840853452682,
|
|
"num_tokens": 5068465.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.360443639755249,
|
|
"epoch": 0.13881582792682215,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00030460198079576355,
|
|
"loss": 6.0397,
|
|
"mean_token_accuracy": 0.14447517544031144,
|
|
"num_tokens": 5079162.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.3138045310974125,
|
|
"epoch": 0.13910807177508913,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003034337235661648,
|
|
"loss": 6.0627,
|
|
"mean_token_accuracy": 0.14073965400457383,
|
|
"num_tokens": 5089409.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.346354389190674,
|
|
"epoch": 0.13940031562335614,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003022646868116714,
|
|
"loss": 6.0073,
|
|
"mean_token_accuracy": 0.15402419716119767,
|
|
"num_tokens": 5100150.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.2642158508300785,
|
|
"epoch": 0.13969255947162312,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003010949025820163,
|
|
"loss": 5.9723,
|
|
"mean_token_accuracy": 0.15449547916650772,
|
|
"num_tokens": 5110601.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.297504043579101,
|
|
"epoch": 0.1399848033198901,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002999244029474252,
|
|
"loss": 6.0562,
|
|
"mean_token_accuracy": 0.1438538871705532,
|
|
"num_tokens": 5119665.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.35467758178711,
|
|
"epoch": 0.14027704716815712,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00029875321999773684,
|
|
"loss": 5.9881,
|
|
"mean_token_accuracy": 0.15138252228498458,
|
|
"num_tokens": 5129712.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.329247283935547,
|
|
"epoch": 0.1405692910164241,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00029758138584152333,
|
|
"loss": 6.0695,
|
|
"mean_token_accuracy": 0.14784711748361587,
|
|
"num_tokens": 5139858.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.434852170944214,
|
|
"epoch": 0.1408615348646911,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002964089326052102,
|
|
"loss": 6.0511,
|
|
"mean_token_accuracy": 0.14410493820905684,
|
|
"num_tokens": 5149330.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.296307039260864,
|
|
"epoch": 0.1411537787129581,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002952358924321949,
|
|
"loss": 6.0336,
|
|
"mean_token_accuracy": 0.15123162269592286,
|
|
"num_tokens": 5159458.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.350845432281494,
|
|
"epoch": 0.1414460225612251,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00029406229748196657,
|
|
"loss": 6.0538,
|
|
"mean_token_accuracy": 0.14705884456634521,
|
|
"num_tokens": 5170439.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.268113899230957,
|
|
"epoch": 0.14173826640949208,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0002928881799292235,
|
|
"loss": 5.9838,
|
|
"mean_token_accuracy": 0.1529225915670395,
|
|
"num_tokens": 5181345.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.332951450347901,
|
|
"epoch": 0.14203051025775906,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00029171357196299154,
|
|
"loss": 6.0038,
|
|
"mean_token_accuracy": 0.1513151541352272,
|
|
"num_tokens": 5191271.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.287766170501709,
|
|
"epoch": 0.14232275410602607,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0002905385057857414,
|
|
"loss": 6.0306,
|
|
"mean_token_accuracy": 0.14540833234786987,
|
|
"num_tokens": 5201868.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.353308391571045,
|
|
"epoch": 0.14261499795429305,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002893630136125058,
|
|
"loss": 6.0348,
|
|
"mean_token_accuracy": 0.14496010169386864,
|
|
"num_tokens": 5212463.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.313853979110718,
|
|
"epoch": 0.14290724180256006,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002881871276699967,
|
|
"loss": 6.0288,
|
|
"mean_token_accuracy": 0.1504661425948143,
|
|
"num_tokens": 5223485.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.290103721618652,
|
|
"epoch": 0.14319948565082705,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00028701088019572114,
|
|
"loss": 5.9775,
|
|
"mean_token_accuracy": 0.15286858677864074,
|
|
"num_tokens": 5233818.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.268964719772339,
|
|
"epoch": 0.14349172949909406,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0002858343034370977,
|
|
"loss": 6.0192,
|
|
"mean_token_accuracy": 0.1612694352865219,
|
|
"num_tokens": 5244008.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.223029232025146,
|
|
"epoch": 0.14378397334736104,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00028465742965057267,
|
|
"loss": 5.9605,
|
|
"mean_token_accuracy": 0.15433412343263625,
|
|
"num_tokens": 5254263.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.358925867080688,
|
|
"epoch": 0.14407621719562802,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00028348029110073533,
|
|
"loss": 6.0514,
|
|
"mean_token_accuracy": 0.14305417686700822,
|
|
"num_tokens": 5264635.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.2891651630401615,
|
|
"epoch": 0.14436846104389503,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00028230292005943365,
|
|
"loss": 6.0385,
|
|
"mean_token_accuracy": 0.15182356610894204,
|
|
"num_tokens": 5275795.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.264105224609375,
|
|
"epoch": 0.144660704892162,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00028112534880488945,
|
|
"loss": 5.9957,
|
|
"mean_token_accuracy": 0.1517108455300331,
|
|
"num_tokens": 5286188.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.31910662651062,
|
|
"epoch": 0.14495294874042902,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002799476096208137,
|
|
"loss": 6.0043,
|
|
"mean_token_accuracy": 0.14954675287008284,
|
|
"num_tokens": 5296224.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.275746059417725,
|
|
"epoch": 0.145245192588696,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00027876973479552087,
|
|
"loss": 6.0083,
|
|
"mean_token_accuracy": 0.15013067051768303,
|
|
"num_tokens": 5308199.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.341499519348145,
|
|
"epoch": 0.145537436436963,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00027759175662104424,
|
|
"loss": 6.0034,
|
|
"mean_token_accuracy": 0.15202850550413133,
|
|
"num_tokens": 5319141.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.36463737487793,
|
|
"epoch": 0.14582968028523,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0002764137073922508,
|
|
"loss": 6.0876,
|
|
"mean_token_accuracy": 0.14759052470326423,
|
|
"num_tokens": 5331387.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.340337610244751,
|
|
"epoch": 0.14612192413349698,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00027523561940595505,
|
|
"loss": 5.9975,
|
|
"mean_token_accuracy": 0.1468089461326599,
|
|
"num_tokens": 5341374.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.3170370101928714,
|
|
"epoch": 0.146414167981764,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002740575249600342,
|
|
"loss": 6.0161,
|
|
"mean_token_accuracy": 0.1478159546852112,
|
|
"num_tokens": 5351728.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.317452669143677,
|
|
"epoch": 0.14670641183003097,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00027287945635254263,
|
|
"loss": 5.9839,
|
|
"mean_token_accuracy": 0.15488136261701585,
|
|
"num_tokens": 5361365.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.265701580047607,
|
|
"epoch": 0.14699865567829798,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00027170144588082635,
|
|
"loss": 6.0273,
|
|
"mean_token_accuracy": 0.1483263149857521,
|
|
"num_tokens": 5372932.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.352311897277832,
|
|
"epoch": 0.14729089952656496,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00027052352584063763,
|
|
"loss": 6.033,
|
|
"mean_token_accuracy": 0.15657648146152497,
|
|
"num_tokens": 5384409.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.288667631149292,
|
|
"epoch": 0.14758314337483197,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00026934572852524907,
|
|
"loss": 5.9337,
|
|
"mean_token_accuracy": 0.15720220953226088,
|
|
"num_tokens": 5394068.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.312388324737549,
|
|
"epoch": 0.14787538722309895,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00026816808622456937,
|
|
"loss": 6.0587,
|
|
"mean_token_accuracy": 0.1438172474503517,
|
|
"num_tokens": 5403908.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.334784841537475,
|
|
"epoch": 0.14816763107136594,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0002669906312242569,
|
|
"loss": 5.9684,
|
|
"mean_token_accuracy": 0.15703339204192163,
|
|
"num_tokens": 5414845.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.273650789260865,
|
|
"epoch": 0.14845987491963294,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00026581339580483525,
|
|
"loss": 5.9692,
|
|
"mean_token_accuracy": 0.14930712282657624,
|
|
"num_tokens": 5424608.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.320654630661011,
|
|
"epoch": 0.14875211876789993,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0002646364122408082,
|
|
"loss": 5.978,
|
|
"mean_token_accuracy": 0.14491919204592704,
|
|
"num_tokens": 5436316.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.355360698699951,
|
|
"epoch": 0.14904436261616694,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002634597127997749,
|
|
"loss": 6.0151,
|
|
"mean_token_accuracy": 0.15137199610471724,
|
|
"num_tokens": 5446112.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.348347330093384,
|
|
"epoch": 0.14933660646443392,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0002622833297415445,
|
|
"loss": 6.0197,
|
|
"mean_token_accuracy": 0.14264217540621757,
|
|
"num_tokens": 5456688.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.227771425247193,
|
|
"epoch": 0.14962885031270093,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002611072953172531,
|
|
"loss": 5.9556,
|
|
"mean_token_accuracy": 0.15220817178487778,
|
|
"num_tokens": 5467590.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.286277532577515,
|
|
"epoch": 0.1499210941609679,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00025993164176847845,
|
|
"loss": 6.0663,
|
|
"mean_token_accuracy": 0.1468990609049797,
|
|
"num_tokens": 5478868.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.387682151794434,
|
|
"epoch": 0.1502133380092349,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002587564013263564,
|
|
"loss": 5.9887,
|
|
"mean_token_accuracy": 0.1580376297235489,
|
|
"num_tokens": 5490024.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.289829730987549,
|
|
"epoch": 0.1505055818575019,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002575816062106974,
|
|
"loss": 6.0214,
|
|
"mean_token_accuracy": 0.15528702437877656,
|
|
"num_tokens": 5500974.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.2699133396148685,
|
|
"epoch": 0.15079782570576888,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00025640728862910293,
|
|
"loss": 5.9653,
|
|
"mean_token_accuracy": 0.1449893519282341,
|
|
"num_tokens": 5511892.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.368878936767578,
|
|
"epoch": 0.1510900695540359,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00025523348077608285,
|
|
"loss": 6.1226,
|
|
"mean_token_accuracy": 0.14047732800245286,
|
|
"num_tokens": 5523913.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.363551235198974,
|
|
"epoch": 0.15138231340230288,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00025406021483217225,
|
|
"loss": 5.9772,
|
|
"mean_token_accuracy": 0.1532949447631836,
|
|
"num_tokens": 5535885.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.308110284805298,
|
|
"epoch": 0.15167455725056989,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00025288752296304963,
|
|
"loss": 6.085,
|
|
"mean_token_accuracy": 0.13639543280005456,
|
|
"num_tokens": 5546087.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.302505922317505,
|
|
"epoch": 0.15196680109883687,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000251715437318655,
|
|
"loss": 6.091,
|
|
"mean_token_accuracy": 0.1428159460425377,
|
|
"num_tokens": 5558295.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.287006664276123,
|
|
"epoch": 0.15225904494710385,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002505439900323084,
|
|
"loss": 5.9298,
|
|
"mean_token_accuracy": 0.15107120499014853,
|
|
"num_tokens": 5568669.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.304688453674316,
|
|
"epoch": 0.15255128879537086,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00024937321321982894,
|
|
"loss": 6.0246,
|
|
"mean_token_accuracy": 0.1511759489774704,
|
|
"num_tokens": 5580134.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.299784326553345,
|
|
"epoch": 0.15284353264363784,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00024820313897865433,
|
|
"loss": 5.961,
|
|
"mean_token_accuracy": 0.15625140368938445,
|
|
"num_tokens": 5589562.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.319279575347901,
|
|
"epoch": 0.15313577649190485,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00024703379938696105,
|
|
"loss": 6.0068,
|
|
"mean_token_accuracy": 0.1511688709259033,
|
|
"num_tokens": 5599944.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.250988578796386,
|
|
"epoch": 0.15342802034017183,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00024586522650278447,
|
|
"loss": 5.9167,
|
|
"mean_token_accuracy": 0.15715879648923875,
|
|
"num_tokens": 5610006.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.250146770477295,
|
|
"epoch": 0.15372026418843884,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00024469745236314064,
|
|
"loss": 5.9862,
|
|
"mean_token_accuracy": 0.15374237895011902,
|
|
"num_tokens": 5620287.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.299891185760498,
|
|
"epoch": 0.15401250803670583,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00024353050898314767,
|
|
"loss": 5.9865,
|
|
"mean_token_accuracy": 0.1511371836066246,
|
|
"num_tokens": 5630614.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.294261789321899,
|
|
"epoch": 0.15430475188497284,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00024236442835514743,
|
|
"loss": 6.0313,
|
|
"mean_token_accuracy": 0.14918752163648605,
|
|
"num_tokens": 5642097.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.2733110904693605,
|
|
"epoch": 0.15459699573323982,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00024119924244782965,
|
|
"loss": 6.0239,
|
|
"mean_token_accuracy": 0.14698346257209777,
|
|
"num_tokens": 5653175.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.312636423110962,
|
|
"epoch": 0.1548892395815068,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00024003498320535462,
|
|
"loss": 5.9992,
|
|
"mean_token_accuracy": 0.15541867315769195,
|
|
"num_tokens": 5663618.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.310853004455566,
|
|
"epoch": 0.1551814834297738,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00023887168254647727,
|
|
"loss": 5.8461,
|
|
"mean_token_accuracy": 0.16287131309509278,
|
|
"num_tokens": 5673460.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.28970890045166,
|
|
"epoch": 0.1554737272780408,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00023770937236367308,
|
|
"loss": 5.9348,
|
|
"mean_token_accuracy": 0.14952913597226142,
|
|
"num_tokens": 5684224.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.310828685760498,
|
|
"epoch": 0.1557659711263078,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00023654808452226278,
|
|
"loss": 5.9855,
|
|
"mean_token_accuracy": 0.15043364316225052,
|
|
"num_tokens": 5694440.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.355057096481323,
|
|
"epoch": 0.15605821497457478,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00023538785085953912,
|
|
"loss": 5.9417,
|
|
"mean_token_accuracy": 0.15679079294204712,
|
|
"num_tokens": 5705510.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.292229461669922,
|
|
"epoch": 0.1563504588228418,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00023422870318389404,
|
|
"loss": 6.0486,
|
|
"mean_token_accuracy": 0.14631869941949843,
|
|
"num_tokens": 5715837.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.294525766372681,
|
|
"epoch": 0.15664270267110877,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002330706732739468,
|
|
"loss": 5.9862,
|
|
"mean_token_accuracy": 0.14949700236320496,
|
|
"num_tokens": 5726252.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.3139581203460695,
|
|
"epoch": 0.15693494651937576,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00023191379287767211,
|
|
"loss": 6.0559,
|
|
"mean_token_accuracy": 0.15128493905067444,
|
|
"num_tokens": 5737877.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.257204580307007,
|
|
"epoch": 0.15722719036764277,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002307580937115305,
|
|
"loss": 5.9708,
|
|
"mean_token_accuracy": 0.14995031505823136,
|
|
"num_tokens": 5748651.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.214744758605957,
|
|
"epoch": 0.15751943421590975,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00022960360745959846,
|
|
"loss": 5.9326,
|
|
"mean_token_accuracy": 0.15505573153495789,
|
|
"num_tokens": 5759505.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.290683221817017,
|
|
"epoch": 0.15781167806417676,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00022845036577269972,
|
|
"loss": 5.9183,
|
|
"mean_token_accuracy": 0.15150227397680283,
|
|
"num_tokens": 5769950.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.317072534561158,
|
|
"epoch": 0.15810392191244374,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00022729840026753777,
|
|
"loss": 5.9447,
|
|
"mean_token_accuracy": 0.15681618750095366,
|
|
"num_tokens": 5780446.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.262074375152588,
|
|
"epoch": 0.15839616576071075,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002261477425258287,
|
|
"loss": 5.9049,
|
|
"mean_token_accuracy": 0.15653743147850036,
|
|
"num_tokens": 5790690.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.276272583007812,
|
|
"epoch": 0.15868840960897773,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002249984240934358,
|
|
"loss": 6.0049,
|
|
"mean_token_accuracy": 0.15046160370111467,
|
|
"num_tokens": 5801718.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.261381149291992,
|
|
"epoch": 0.1589806534572447,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00022385047647950464,
|
|
"loss": 5.9795,
|
|
"mean_token_accuracy": 0.15426528006792067,
|
|
"num_tokens": 5812497.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.295159435272216,
|
|
"epoch": 0.15927289730551172,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002227039311555986,
|
|
"loss": 5.9623,
|
|
"mean_token_accuracy": 0.15989028364419938,
|
|
"num_tokens": 5822931.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.30918321609497,
|
|
"epoch": 0.1595651411537787,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002215588195548372,
|
|
"loss": 5.9153,
|
|
"mean_token_accuracy": 0.15595535635948182,
|
|
"num_tokens": 5832942.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.281555843353272,
|
|
"epoch": 0.15985738500204572,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00022041517307103337,
|
|
"loss": 5.9554,
|
|
"mean_token_accuracy": 0.14854271337389946,
|
|
"num_tokens": 5842830.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.257166862487793,
|
|
"epoch": 0.1601496288503127,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0002192730230578331,
|
|
"loss": 5.9631,
|
|
"mean_token_accuracy": 0.14914022982120514,
|
|
"num_tokens": 5854392.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.352614259719848,
|
|
"epoch": 0.1604418726985797,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002181324008278559,
|
|
"loss": 5.973,
|
|
"mean_token_accuracy": 0.1517603389918804,
|
|
"num_tokens": 5865582.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.298333597183228,
|
|
"epoch": 0.1607341165468467,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00021699333765183655,
|
|
"loss": 5.9648,
|
|
"mean_token_accuracy": 0.1535767138004303,
|
|
"num_tokens": 5875669.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.270532560348511,
|
|
"epoch": 0.16102636039511367,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002158558647577673,
|
|
"loss": 5.9193,
|
|
"mean_token_accuracy": 0.1563738539814949,
|
|
"num_tokens": 5886287.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.26015796661377,
|
|
"epoch": 0.16131860424338068,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00021472001333004215,
|
|
"loss": 5.954,
|
|
"mean_token_accuracy": 0.15786738842725753,
|
|
"num_tokens": 5896884.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.220677423477173,
|
|
"epoch": 0.16161084809164766,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00021358581450860186,
|
|
"loss": 6.0411,
|
|
"mean_token_accuracy": 0.14786829352378844,
|
|
"num_tokens": 5907741.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.383463573455811,
|
|
"epoch": 0.16190309193991467,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0002124532993880799,
|
|
"loss": 6.0218,
|
|
"mean_token_accuracy": 0.15272231251001359,
|
|
"num_tokens": 5918316.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.325487995147705,
|
|
"epoch": 0.16219533578818165,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00021132249901695044,
|
|
"loss": 6.0119,
|
|
"mean_token_accuracy": 0.15464432835578917,
|
|
"num_tokens": 5929209.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.295078992843628,
|
|
"epoch": 0.16248757963644866,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00021019344439667705,
|
|
"loss": 5.9468,
|
|
"mean_token_accuracy": 0.15779951959848404,
|
|
"num_tokens": 5939716.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.254607343673706,
|
|
"epoch": 0.16277982348471565,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00020906616648086213,
|
|
"loss": 5.9041,
|
|
"mean_token_accuracy": 0.1592554844915867,
|
|
"num_tokens": 5950677.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.295980501174927,
|
|
"epoch": 0.16307206733298263,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00020794069617439942,
|
|
"loss": 5.9352,
|
|
"mean_token_accuracy": 0.15921037048101425,
|
|
"num_tokens": 5961458.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.256237316131592,
|
|
"epoch": 0.16336431118124964,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00020681706433262593,
|
|
"loss": 5.9022,
|
|
"mean_token_accuracy": 0.15727142244577408,
|
|
"num_tokens": 5971713.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.305441045761109,
|
|
"epoch": 0.16365655502951662,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00020569530176047602,
|
|
"loss": 6.0447,
|
|
"mean_token_accuracy": 0.14972822219133378,
|
|
"num_tokens": 5983418.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.34089150428772,
|
|
"epoch": 0.16394879887778363,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002045754392116374,
|
|
"loss": 5.869,
|
|
"mean_token_accuracy": 0.16185261458158492,
|
|
"num_tokens": 5993207.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.237485265731811,
|
|
"epoch": 0.1642410427260506,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00020345750738770757,
|
|
"loss": 5.9724,
|
|
"mean_token_accuracy": 0.14559953808784484,
|
|
"num_tokens": 6003545.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.2489461421966555,
|
|
"epoch": 0.16453328657431762,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00020234153693735214,
|
|
"loss": 5.9876,
|
|
"mean_token_accuracy": 0.14853048473596572,
|
|
"num_tokens": 6014315.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.331522417068482,
|
|
"epoch": 0.1648255304225846,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0002012275584554647,
|
|
"loss": 5.879,
|
|
"mean_token_accuracy": 0.15474118292331696,
|
|
"num_tokens": 6025650.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.312690925598145,
|
|
"epoch": 0.16511777427085159,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00020011560248232803,
|
|
"loss": 5.9071,
|
|
"mean_token_accuracy": 0.15482232123613357,
|
|
"num_tokens": 6036497.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.223179769515991,
|
|
"epoch": 0.1654100181191186,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00019900569950277692,
|
|
"loss": 5.8929,
|
|
"mean_token_accuracy": 0.15263683944940568,
|
|
"num_tokens": 6046448.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.252099275588989,
|
|
"epoch": 0.16570226196738558,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00019789787994536228,
|
|
"loss": 5.88,
|
|
"mean_token_accuracy": 0.15355804860591887,
|
|
"num_tokens": 6056630.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.260057497024536,
|
|
"epoch": 0.1659945058156526,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00019679217418151667,
|
|
"loss": 5.937,
|
|
"mean_token_accuracy": 0.15121540874242784,
|
|
"num_tokens": 6067795.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.27913122177124,
|
|
"epoch": 0.16628674966391957,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00019568861252472236,
|
|
"loss": 5.9489,
|
|
"mean_token_accuracy": 0.16291055530309678,
|
|
"num_tokens": 6079218.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.242688941955566,
|
|
"epoch": 0.16657899351218658,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00019458722522967952,
|
|
"loss": 5.9344,
|
|
"mean_token_accuracy": 0.15513963103294373,
|
|
"num_tokens": 6089707.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.2319025039672855,
|
|
"epoch": 0.16687123736045356,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00019348804249147723,
|
|
"loss": 5.9286,
|
|
"mean_token_accuracy": 0.15406835228204727,
|
|
"num_tokens": 6100978.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.288013601303101,
|
|
"epoch": 0.16716348120872054,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001923910944447655,
|
|
"loss": 5.8506,
|
|
"mean_token_accuracy": 0.1609133541584015,
|
|
"num_tokens": 6110188.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.294746685028076,
|
|
"epoch": 0.16745572505698755,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00019129641116292928,
|
|
"loss": 5.9571,
|
|
"mean_token_accuracy": 0.15872167944908142,
|
|
"num_tokens": 6121229.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.239352464675903,
|
|
"epoch": 0.16774796890525454,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00019020402265726343,
|
|
"loss": 5.9593,
|
|
"mean_token_accuracy": 0.15123945772647857,
|
|
"num_tokens": 6131022.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.177235221862793,
|
|
"epoch": 0.16804021275352155,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001891139588761509,
|
|
"loss": 5.8541,
|
|
"mean_token_accuracy": 0.16396355330944062,
|
|
"num_tokens": 6141042.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.314100122451782,
|
|
"epoch": 0.16833245660178853,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00018802624970424076,
|
|
"loss": 5.9393,
|
|
"mean_token_accuracy": 0.15475943088531494,
|
|
"num_tokens": 6151612.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.2800627708435055,
|
|
"epoch": 0.16862470045005554,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00018694092496162945,
|
|
"loss": 5.937,
|
|
"mean_token_accuracy": 0.1557922825217247,
|
|
"num_tokens": 6162217.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.312515592575073,
|
|
"epoch": 0.16891694429832252,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00018585801440304306,
|
|
"loss": 5.9766,
|
|
"mean_token_accuracy": 0.1531226873397827,
|
|
"num_tokens": 6173309.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.311217832565307,
|
|
"epoch": 0.1692091881465895,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00018477754771702165,
|
|
"loss": 5.9567,
|
|
"mean_token_accuracy": 0.1584221139550209,
|
|
"num_tokens": 6184063.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.245030498504638,
|
|
"epoch": 0.1695014319948565,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00018369955452510506,
|
|
"loss": 5.9231,
|
|
"mean_token_accuracy": 0.1564079113304615,
|
|
"num_tokens": 6196158.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.233873271942139,
|
|
"epoch": 0.1697936758431235,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001826240643810212,
|
|
"loss": 5.9005,
|
|
"mean_token_accuracy": 0.15264101922512055,
|
|
"num_tokens": 6206801.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.284539604187012,
|
|
"epoch": 0.1700859196913905,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001815511067698758,
|
|
"loss": 5.9452,
|
|
"mean_token_accuracy": 0.15421261191368102,
|
|
"num_tokens": 6217060.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.290277528762817,
|
|
"epoch": 0.17037816353965748,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001804807111073436,
|
|
"loss": 5.9119,
|
|
"mean_token_accuracy": 0.15774459540843963,
|
|
"num_tokens": 6228311.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.276092576980591,
|
|
"epoch": 0.1706704073879245,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001794129067388625,
|
|
"loss": 5.9323,
|
|
"mean_token_accuracy": 0.15433253049850465,
|
|
"num_tokens": 6238059.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.230207300186157,
|
|
"epoch": 0.17096265123619148,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00017834772293882868,
|
|
"loss": 5.9629,
|
|
"mean_token_accuracy": 0.15217517614364623,
|
|
"num_tokens": 6248462.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.255596065521241,
|
|
"epoch": 0.17125489508445846,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000177285188909794,
|
|
"loss": 5.872,
|
|
"mean_token_accuracy": 0.16722603738307953,
|
|
"num_tokens": 6257974.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.335262107849121,
|
|
"epoch": 0.17154713893272547,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0001762253337816656,
|
|
"loss": 5.9814,
|
|
"mean_token_accuracy": 0.1485115647315979,
|
|
"num_tokens": 6269296.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.358278417587281,
|
|
"epoch": 0.17183938278099245,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00017516818661090738,
|
|
"loss": 5.9916,
|
|
"mean_token_accuracy": 0.15062511041760446,
|
|
"num_tokens": 6280104.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.311861419677735,
|
|
"epoch": 0.17213162662925946,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001741137763797428,
|
|
"loss": 5.9015,
|
|
"mean_token_accuracy": 0.1671626940369606,
|
|
"num_tokens": 6290477.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.258217287063599,
|
|
"epoch": 0.17242387047752644,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00017306213199536115,
|
|
"loss": 5.8602,
|
|
"mean_token_accuracy": 0.16427712887525558,
|
|
"num_tokens": 6300927.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.236834526062012,
|
|
"epoch": 0.17271611432579345,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0001720132822891243,
|
|
"loss": 5.9459,
|
|
"mean_token_accuracy": 0.15635040700435637,
|
|
"num_tokens": 6311153.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.292562246322632,
|
|
"epoch": 0.17300835817406043,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001709672560157769,
|
|
"loss": 5.8447,
|
|
"mean_token_accuracy": 0.16232619136571885,
|
|
"num_tokens": 6320716.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.313056468963623,
|
|
"epoch": 0.17330060202232744,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00016992408185265758,
|
|
"loss": 6.0058,
|
|
"mean_token_accuracy": 0.14952416867017745,
|
|
"num_tokens": 6332432.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.296169853210449,
|
|
"epoch": 0.17359284587059443,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00016888378839891298,
|
|
"loss": 5.9592,
|
|
"mean_token_accuracy": 0.14839272052049637,
|
|
"num_tokens": 6343577.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.259261178970337,
|
|
"epoch": 0.1738850897188614,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0001678464041747137,
|
|
"loss": 5.9651,
|
|
"mean_token_accuracy": 0.15139905959367753,
|
|
"num_tokens": 6353689.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.2840352058410645,
|
|
"epoch": 0.17417733356712842,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00016681195762047223,
|
|
"loss": 5.9491,
|
|
"mean_token_accuracy": 0.15322353988885878,
|
|
"num_tokens": 6365075.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.284506511688233,
|
|
"epoch": 0.1744695774153954,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00016578047709606337,
|
|
"loss": 5.9137,
|
|
"mean_token_accuracy": 0.16487067937850952,
|
|
"num_tokens": 6375189.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.261584281921387,
|
|
"epoch": 0.1747618212636624,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00016475199088004678,
|
|
"loss": 5.9027,
|
|
"mean_token_accuracy": 0.15671227872371674,
|
|
"num_tokens": 6385028.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.270907258987426,
|
|
"epoch": 0.1750540651119294,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00016372652716889163,
|
|
"loss": 5.9415,
|
|
"mean_token_accuracy": 0.14871460497379302,
|
|
"num_tokens": 6395912.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.318777847290039,
|
|
"epoch": 0.1753463089601964,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001627041140762035,
|
|
"loss": 5.8682,
|
|
"mean_token_accuracy": 0.15546663701534272,
|
|
"num_tokens": 6405993.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 6.283399152755737,
|
|
"epoch": 0.17563855280846338,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00016168477963195382,
|
|
"loss": 5.9756,
|
|
"mean_token_accuracy": 0.1521611973643303,
|
|
"num_tokens": 6417085.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 6.2866161346435545,
|
|
"epoch": 0.17593079665673036,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001606685517817114,
|
|
"loss": 5.8998,
|
|
"mean_token_accuracy": 0.15176964104175567,
|
|
"num_tokens": 6426768.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 6.262517881393433,
|
|
"epoch": 0.17622304050499737,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00015965545838587592,
|
|
"loss": 5.9442,
|
|
"mean_token_accuracy": 0.16012927442789077,
|
|
"num_tokens": 6437068.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 6.298662185668945,
|
|
"epoch": 0.17651528435326436,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00015864552721891467,
|
|
"loss": 5.9239,
|
|
"mean_token_accuracy": 0.1625131204724312,
|
|
"num_tokens": 6447480.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.237483263015747,
|
|
"epoch": 0.17680752820153137,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00015763878596860076,
|
|
"loss": 5.9547,
|
|
"mean_token_accuracy": 0.15520766079425813,
|
|
"num_tokens": 6458014.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 6.32170672416687,
|
|
"epoch": 0.17709977204979835,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00015663526223525412,
|
|
"loss": 5.9367,
|
|
"mean_token_accuracy": 0.15889369696378708,
|
|
"num_tokens": 6467899.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 6.276149654388428,
|
|
"epoch": 0.17739201589806536,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0001556349835309848,
|
|
"loss": 5.9776,
|
|
"mean_token_accuracy": 0.14879269748926163,
|
|
"num_tokens": 6480192.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 6.2329589366912845,
|
|
"epoch": 0.17768425974633234,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0001546379772789389,
|
|
"loss": 5.9551,
|
|
"mean_token_accuracy": 0.15424026623368264,
|
|
"num_tokens": 6492019.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.322043323516846,
|
|
"epoch": 0.17797650359459932,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00015364427081254622,
|
|
"loss": 5.8893,
|
|
"mean_token_accuracy": 0.16130917072296141,
|
|
"num_tokens": 6503478.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 6.256545495986939,
|
|
"epoch": 0.17826874744286633,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00015265389137477165,
|
|
"loss": 5.8816,
|
|
"mean_token_accuracy": 0.15927795171737671,
|
|
"num_tokens": 6513982.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 6.287691116333008,
|
|
"epoch": 0.17856099129113331,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00015166686611736786,
|
|
"loss": 5.8846,
|
|
"mean_token_accuracy": 0.1545673280954361,
|
|
"num_tokens": 6524798.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 6.238642454147339,
|
|
"epoch": 0.17885323513940032,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00015068322210013064,
|
|
"loss": 5.8386,
|
|
"mean_token_accuracy": 0.1584905982017517,
|
|
"num_tokens": 6536105.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 6.26782922744751,
|
|
"epoch": 0.1791454789876673,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0001497029862901578,
|
|
"loss": 5.9609,
|
|
"mean_token_accuracy": 0.15500572472810745,
|
|
"num_tokens": 6547250.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 6.257089996337891,
|
|
"epoch": 0.17943772283593432,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00014872618556110905,
|
|
"loss": 5.8859,
|
|
"mean_token_accuracy": 0.1602465182542801,
|
|
"num_tokens": 6556990.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 6.1768107414245605,
|
|
"epoch": 0.1797299666842013,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00014775284669246992,
|
|
"loss": 5.8949,
|
|
"mean_token_accuracy": 0.15843170285224914,
|
|
"num_tokens": 6567916.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 6.210200643539428,
|
|
"epoch": 0.18002221053246828,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00014678299636881716,
|
|
"loss": 5.8162,
|
|
"mean_token_accuracy": 0.16867512762546538,
|
|
"num_tokens": 6578775.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 6.217078065872192,
|
|
"epoch": 0.1803144543807353,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001458166611790873,
|
|
"loss": 5.8867,
|
|
"mean_token_accuracy": 0.16877478063106538,
|
|
"num_tokens": 6588779.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 6.325368118286133,
|
|
"epoch": 0.18060669822900227,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00014485386761584773,
|
|
"loss": 5.9561,
|
|
"mean_token_accuracy": 0.15594409108161927,
|
|
"num_tokens": 6599315.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.343292951583862,
|
|
"epoch": 0.18089894207726928,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00014389464207457042,
|
|
"loss": 5.977,
|
|
"mean_token_accuracy": 0.1541369929909706,
|
|
"num_tokens": 6610900.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 6.318844509124756,
|
|
"epoch": 0.18119118592553626,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00014293901085290795,
|
|
"loss": 5.9314,
|
|
"mean_token_accuracy": 0.1589741140604019,
|
|
"num_tokens": 6621888.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 6.229228878021241,
|
|
"epoch": 0.18148342977380327,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00014198700014997307,
|
|
"loss": 5.9603,
|
|
"mean_token_accuracy": 0.1509167268872261,
|
|
"num_tokens": 6632228.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 6.231994199752807,
|
|
"epoch": 0.18177567362207026,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00014103863606562016,
|
|
"loss": 5.8834,
|
|
"mean_token_accuracy": 0.16131148487329483,
|
|
"num_tokens": 6642354.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 6.221903371810913,
|
|
"epoch": 0.18206791747033724,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00014009394459972964,
|
|
"loss": 5.8589,
|
|
"mean_token_accuracy": 0.16473948806524277,
|
|
"num_tokens": 6653025.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 6.255872964859009,
|
|
"epoch": 0.18236016131860425,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00013915295165149513,
|
|
"loss": 5.8858,
|
|
"mean_token_accuracy": 0.1564686566591263,
|
|
"num_tokens": 6663326.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 6.3000284194946286,
|
|
"epoch": 0.18265240516687123,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00013821568301871384,
|
|
"loss": 5.9204,
|
|
"mean_token_accuracy": 0.15571961998939515,
|
|
"num_tokens": 6674104.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 6.257794427871704,
|
|
"epoch": 0.18294464901513824,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00013728216439707862,
|
|
"loss": 5.8444,
|
|
"mean_token_accuracy": 0.16243265718221664,
|
|
"num_tokens": 6684442.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 6.297143602371216,
|
|
"epoch": 0.18323689286340522,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00013635242137947419,
|
|
"loss": 5.9649,
|
|
"mean_token_accuracy": 0.15052273720502854,
|
|
"num_tokens": 6695683.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 6.2341681003570555,
|
|
"epoch": 0.18352913671167223,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00013542647945527498,
|
|
"loss": 5.8073,
|
|
"mean_token_accuracy": 0.16549235731363296,
|
|
"num_tokens": 6705252.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 6.2586750984191895,
|
|
"epoch": 0.1838213805599392,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001345043640096465,
|
|
"loss": 5.8929,
|
|
"mean_token_accuracy": 0.1568222850561142,
|
|
"num_tokens": 6715800.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.275263833999634,
|
|
"epoch": 0.1841136244082062,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00013358610032284957,
|
|
"loss": 5.827,
|
|
"mean_token_accuracy": 0.15784039795398713,
|
|
"num_tokens": 6726027.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 6.254611206054688,
|
|
"epoch": 0.1844058682564732,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000132671713569547,
|
|
"loss": 5.8536,
|
|
"mean_token_accuracy": 0.15593142211437225,
|
|
"num_tokens": 6736023.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 6.220268058776855,
|
|
"epoch": 0.1846981121047402,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001317612288181136,
|
|
"loss": 5.9378,
|
|
"mean_token_accuracy": 0.15204131305217744,
|
|
"num_tokens": 6747433.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 6.2448375701904295,
|
|
"epoch": 0.1849903559530072,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00013085467102994864,
|
|
"loss": 5.8586,
|
|
"mean_token_accuracy": 0.15780818611383438,
|
|
"num_tokens": 6758207.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 6.2656317234039305,
|
|
"epoch": 0.18528259980127418,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00012995206505879198,
|
|
"loss": 5.8944,
|
|
"mean_token_accuracy": 0.16099692285060882,
|
|
"num_tokens": 6768640.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.305795383453369,
|
|
"epoch": 0.1855748436495412,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001290534356500421,
|
|
"loss": 5.878,
|
|
"mean_token_accuracy": 0.1598610058426857,
|
|
"num_tokens": 6780244.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 6.227605533599854,
|
|
"epoch": 0.18586708749780817,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00012815880744007827,
|
|
"loss": 5.9234,
|
|
"mean_token_accuracy": 0.15409687906503677,
|
|
"num_tokens": 6790334.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 6.286476898193359,
|
|
"epoch": 0.18615933134607515,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00012726820495558483,
|
|
"loss": 5.9309,
|
|
"mean_token_accuracy": 0.15614654123783112,
|
|
"num_tokens": 6800624.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 6.283801078796387,
|
|
"epoch": 0.18645157519434216,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00012638165261287868,
|
|
"loss": 5.9173,
|
|
"mean_token_accuracy": 0.15313496440649033,
|
|
"num_tokens": 6812309.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.247753429412842,
|
|
"epoch": 0.18674381904260914,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001254991747172402,
|
|
"loss": 5.8264,
|
|
"mean_token_accuracy": 0.16600472778081893,
|
|
"num_tokens": 6822699.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 6.194558906555176,
|
|
"epoch": 0.18703606289087615,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00012462079546224662,
|
|
"loss": 5.8231,
|
|
"mean_token_accuracy": 0.16591012626886367,
|
|
"num_tokens": 6832901.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 6.248186922073364,
|
|
"epoch": 0.18732830673914314,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00012374653892910896,
|
|
"loss": 5.9326,
|
|
"mean_token_accuracy": 0.15820514261722565,
|
|
"num_tokens": 6843547.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 6.250892066955567,
|
|
"epoch": 0.18762055058741015,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00012287642908601166,
|
|
"loss": 5.8263,
|
|
"mean_token_accuracy": 0.16346440464258194,
|
|
"num_tokens": 6853779.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.2538402557373045,
|
|
"epoch": 0.18791279443567713,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00012201048978745569,
|
|
"loss": 5.8144,
|
|
"mean_token_accuracy": 0.1630478873848915,
|
|
"num_tokens": 6864417.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 6.2370952606201175,
|
|
"epoch": 0.1882050382839441,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00012114874477360427,
|
|
"loss": 5.8815,
|
|
"mean_token_accuracy": 0.15915295779705046,
|
|
"num_tokens": 6875310.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 6.269949960708618,
|
|
"epoch": 0.18849728213221112,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00012029121766963236,
|
|
"loss": 5.9562,
|
|
"mean_token_accuracy": 0.14962919652462006,
|
|
"num_tokens": 6886732.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 6.281165885925293,
|
|
"epoch": 0.1887895259804781,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00011943793198507858,
|
|
"loss": 5.8957,
|
|
"mean_token_accuracy": 0.15709118247032167,
|
|
"num_tokens": 6897254.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 6.316904973983765,
|
|
"epoch": 0.1890817698287451,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00011858891111320104,
|
|
"loss": 5.9125,
|
|
"mean_token_accuracy": 0.1573660343885422,
|
|
"num_tokens": 6907382.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 6.245898056030273,
|
|
"epoch": 0.1893740136770121,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001177441783303359,
|
|
"loss": 5.9008,
|
|
"mean_token_accuracy": 0.15485315173864364,
|
|
"num_tokens": 6917189.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 6.22566294670105,
|
|
"epoch": 0.1896662575252791,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00011690375679525896,
|
|
"loss": 5.8544,
|
|
"mean_token_accuracy": 0.15707467794418334,
|
|
"num_tokens": 6927523.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 6.285468244552613,
|
|
"epoch": 0.18995850137354608,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00011606766954855124,
|
|
"loss": 5.8779,
|
|
"mean_token_accuracy": 0.1606733500957489,
|
|
"num_tokens": 6939475.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 6.271123743057251,
|
|
"epoch": 0.1902507452218131,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00011523593951196702,
|
|
"loss": 5.8547,
|
|
"mean_token_accuracy": 0.15778176635503768,
|
|
"num_tokens": 6948928.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 6.2798747539520265,
|
|
"epoch": 0.19054298907008008,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00011440858948780523,
|
|
"loss": 5.9219,
|
|
"mean_token_accuracy": 0.1565907746553421,
|
|
"num_tokens": 6959917.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 6.232462787628174,
|
|
"epoch": 0.19083523291834706,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00011358564215828484,
|
|
"loss": 5.8712,
|
|
"mean_token_accuracy": 0.15686557441949844,
|
|
"num_tokens": 6970426.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 6.255841636657715,
|
|
"epoch": 0.19112747676661407,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00011276712008492254,
|
|
"loss": 5.8603,
|
|
"mean_token_accuracy": 0.15819348394870758,
|
|
"num_tokens": 6981412.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 6.236334943771363,
|
|
"epoch": 0.19141972061488105,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00011195304570791451,
|
|
"loss": 5.8288,
|
|
"mean_token_accuracy": 0.16009365618228913,
|
|
"num_tokens": 6992384.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.254592370986939,
|
|
"epoch": 0.19171196446314806,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00011114344134552094,
|
|
"loss": 5.8869,
|
|
"mean_token_accuracy": 0.15505520552396773,
|
|
"num_tokens": 7004033.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 6.225855112075806,
|
|
"epoch": 0.19200420831141504,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001103383291934545,
|
|
"loss": 5.8348,
|
|
"mean_token_accuracy": 0.16033557802438736,
|
|
"num_tokens": 7014473.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 6.279972171783447,
|
|
"epoch": 0.19229645215968205,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00010953773132427141,
|
|
"loss": 5.8444,
|
|
"mean_token_accuracy": 0.1567966401576996,
|
|
"num_tokens": 7024933.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 6.281522560119629,
|
|
"epoch": 0.19258869600794903,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00010874166968676677,
|
|
"loss": 5.8781,
|
|
"mean_token_accuracy": 0.1557956889271736,
|
|
"num_tokens": 7035709.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 6.24364743232727,
|
|
"epoch": 0.19288093985621602,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00010795016610537251,
|
|
"loss": 5.9007,
|
|
"mean_token_accuracy": 0.15638068467378616,
|
|
"num_tokens": 7046090.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 6.247688102722168,
|
|
"epoch": 0.19317318370448303,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00010716324227955904,
|
|
"loss": 5.9125,
|
|
"mean_token_accuracy": 0.15485986471176147,
|
|
"num_tokens": 7057984.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.260491275787354,
|
|
"epoch": 0.19346542755275,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001063809197832406,
|
|
"loss": 5.839,
|
|
"mean_token_accuracy": 0.16005982756614684,
|
|
"num_tokens": 7069717.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 6.289077711105347,
|
|
"epoch": 0.19375767140101702,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00010560322006418368,
|
|
"loss": 5.9414,
|
|
"mean_token_accuracy": 0.1482711210846901,
|
|
"num_tokens": 7080763.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 6.228817510604858,
|
|
"epoch": 0.194049915249284,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00010483016444341887,
|
|
"loss": 5.8543,
|
|
"mean_token_accuracy": 0.16758138090372085,
|
|
"num_tokens": 7090765.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 6.282393932342529,
|
|
"epoch": 0.194342159097551,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00010406177411465654,
|
|
"loss": 5.9164,
|
|
"mean_token_accuracy": 0.16024034470319748,
|
|
"num_tokens": 7101056.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 6.282049322128296,
|
|
"epoch": 0.194634402945818,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00010329807014370562,
|
|
"loss": 5.8344,
|
|
"mean_token_accuracy": 0.16477701663970948,
|
|
"num_tokens": 7112077.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 6.23296799659729,
|
|
"epoch": 0.19492664679408497,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00010253907346789632,
|
|
"loss": 5.7806,
|
|
"mean_token_accuracy": 0.16275889575481414,
|
|
"num_tokens": 7122760.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 6.233717155456543,
|
|
"epoch": 0.19521889064235198,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00010178480489550596,
|
|
"loss": 5.9269,
|
|
"mean_token_accuracy": 0.16597806364297868,
|
|
"num_tokens": 7133883.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 6.250077867507935,
|
|
"epoch": 0.19551113449061897,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00010103528510518836,
|
|
"loss": 5.8742,
|
|
"mean_token_accuracy": 0.15735647082328796,
|
|
"num_tokens": 7143890.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 6.238697147369384,
|
|
"epoch": 0.19580337833888597,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001002905346454073,
|
|
"loss": 5.82,
|
|
"mean_token_accuracy": 0.16303645372390746,
|
|
"num_tokens": 7154256.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 6.26402530670166,
|
|
"epoch": 0.19609562218715296,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 9.955057393387285e-05,
|
|
"loss": 5.873,
|
|
"mean_token_accuracy": 0.16090958714485168,
|
|
"num_tokens": 7165126.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 6.276232147216797,
|
|
"epoch": 0.19638786603541997,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 9.88154232569816e-05,
|
|
"loss": 5.7926,
|
|
"mean_token_accuracy": 0.16471965312957765,
|
|
"num_tokens": 7176532.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 6.237003183364868,
|
|
"epoch": 0.19668010988368695,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 9.808510276926075e-05,
|
|
"loss": 5.8243,
|
|
"mean_token_accuracy": 0.16415449380874633,
|
|
"num_tokens": 7188181.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.213815784454345,
|
|
"epoch": 0.19697235373195393,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 9.735963249281549e-05,
|
|
"loss": 5.7948,
|
|
"mean_token_accuracy": 0.16533007472753525,
|
|
"num_tokens": 7199081.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 6.21927318572998,
|
|
"epoch": 0.19726459758022094,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 9.663903231677974e-05,
|
|
"loss": 5.7777,
|
|
"mean_token_accuracy": 0.16500383913516997,
|
|
"num_tokens": 7209165.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 6.218824005126953,
|
|
"epoch": 0.19755684142848792,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 9.592332199677145e-05,
|
|
"loss": 5.8774,
|
|
"mean_token_accuracy": 0.1631125494837761,
|
|
"num_tokens": 7219973.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 6.241815805435181,
|
|
"epoch": 0.19784908527675493,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 9.521252115435061e-05,
|
|
"loss": 5.8447,
|
|
"mean_token_accuracy": 0.16132166087627411,
|
|
"num_tokens": 7230846.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 6.213486671447754,
|
|
"epoch": 0.19814132912502191,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 9.450664927648126e-05,
|
|
"loss": 5.8499,
|
|
"mean_token_accuracy": 0.1634298399090767,
|
|
"num_tokens": 7241874.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 6.246858358383179,
|
|
"epoch": 0.19843357297328892,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 9.380572571499758e-05,
|
|
"loss": 5.8527,
|
|
"mean_token_accuracy": 0.16297064274549483,
|
|
"num_tokens": 7252468.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 6.28117094039917,
|
|
"epoch": 0.1987258168215559,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 9.310976968607307e-05,
|
|
"loss": 5.9974,
|
|
"mean_token_accuracy": 0.15045272409915925,
|
|
"num_tokens": 7264221.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.245209884643555,
|
|
"epoch": 0.1990180606698229,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 9.241880026969381e-05,
|
|
"loss": 5.8651,
|
|
"mean_token_accuracy": 0.1604456678032875,
|
|
"num_tokens": 7275063.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 6.268118858337402,
|
|
"epoch": 0.1993103045180899,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 9.173283640913537e-05,
|
|
"loss": 5.8696,
|
|
"mean_token_accuracy": 0.15584716200828552,
|
|
"num_tokens": 7285343.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 6.275067472457886,
|
|
"epoch": 0.19960254836635688,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 9.10518969104436e-05,
|
|
"loss": 5.9048,
|
|
"mean_token_accuracy": 0.1594344735145569,
|
|
"num_tokens": 7297296.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 6.2276551723480225,
|
|
"epoch": 0.1998947922146239,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 9.037600044191868e-05,
|
|
"loss": 5.889,
|
|
"mean_token_accuracy": 0.15555773824453353,
|
|
"num_tokens": 7308216.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 6.2653766632080075,
|
|
"epoch": 0.20018703606289087,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 8.970516553360383e-05,
|
|
"loss": 5.8122,
|
|
"mean_token_accuracy": 0.16367650181055068,
|
|
"num_tokens": 7318708.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 6.279593753814697,
|
|
"epoch": 0.20047927991115788,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 8.903941057677692e-05,
|
|
"loss": 5.887,
|
|
"mean_token_accuracy": 0.15948577970266342,
|
|
"num_tokens": 7328989.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 6.302930688858032,
|
|
"epoch": 0.20077152375942486,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 8.837875382344635e-05,
|
|
"loss": 5.9699,
|
|
"mean_token_accuracy": 0.159020034968853,
|
|
"num_tokens": 7340188.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.257382011413574,
|
|
"epoch": 0.20106376760769185,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 8.772321338585076e-05,
|
|
"loss": 5.8884,
|
|
"mean_token_accuracy": 0.15751360058784486,
|
|
"num_tokens": 7350847.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 6.2526270866394045,
|
|
"epoch": 0.20135601145595886,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 8.707280723596242e-05,
|
|
"loss": 5.8704,
|
|
"mean_token_accuracy": 0.16057247072458267,
|
|
"num_tokens": 7361499.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 6.221743869781494,
|
|
"epoch": 0.20164825530422584,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 8.64275532049944e-05,
|
|
"loss": 5.8203,
|
|
"mean_token_accuracy": 0.168203441798687,
|
|
"num_tokens": 7372395.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 6.2632581233978275,
|
|
"epoch": 0.20194049915249285,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 8.578746898291198e-05,
|
|
"loss": 5.8997,
|
|
"mean_token_accuracy": 0.15661046504974366,
|
|
"num_tokens": 7382462.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 6.325605487823486,
|
|
"epoch": 0.20223274300075983,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 8.515257211794742e-05,
|
|
"loss": 5.8138,
|
|
"mean_token_accuracy": 0.16082889437675477,
|
|
"num_tokens": 7392000.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 6.280700826644898,
|
|
"epoch": 0.20252498684902684,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 8.452288001611896e-05,
|
|
"loss": 5.8843,
|
|
"mean_token_accuracy": 0.1621067613363266,
|
|
"num_tokens": 7403286.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 6.2394922256469725,
|
|
"epoch": 0.20281723069729382,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 8.389840994075379e-05,
|
|
"loss": 5.8736,
|
|
"mean_token_accuracy": 0.15633028000593185,
|
|
"num_tokens": 7414412.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 6.266023111343384,
|
|
"epoch": 0.2031094745455608,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 8.327917901201435e-05,
|
|
"loss": 5.8402,
|
|
"mean_token_accuracy": 0.15468731224536897,
|
|
"num_tokens": 7425218.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 6.302168655395508,
|
|
"epoch": 0.2034017183938278,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 8.266520420642931e-05,
|
|
"loss": 5.8598,
|
|
"mean_token_accuracy": 0.16676196753978728,
|
|
"num_tokens": 7435899.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 6.301884126663208,
|
|
"epoch": 0.2036939622420948,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 8.205650235642828e-05,
|
|
"loss": 5.9184,
|
|
"mean_token_accuracy": 0.15453746318817138,
|
|
"num_tokens": 7446812.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 6.283580780029297,
|
|
"epoch": 0.2039862060903618,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 8.145309014987978e-05,
|
|
"loss": 5.9471,
|
|
"mean_token_accuracy": 0.15718840509653093,
|
|
"num_tokens": 7457468.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 6.2381792068481445,
|
|
"epoch": 0.2042784499386288,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 8.085498412963437e-05,
|
|
"loss": 5.8541,
|
|
"mean_token_accuracy": 0.16251549422740935,
|
|
"num_tokens": 7468604.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 6.2204193592071535,
|
|
"epoch": 0.2045706937868958,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 8.026220069307078e-05,
|
|
"loss": 5.8658,
|
|
"mean_token_accuracy": 0.15938508212566377,
|
|
"num_tokens": 7479548.0,
|
|
"step": 3500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"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": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1647496421867520.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|