5177 lines
147 KiB
JSON
5177 lines
147 KiB
JSON
{
|
|
"best_global_step": 512,
|
|
"best_metric": 0.5158528089523315,
|
|
"best_model_checkpoint": "Disctil-Qwen3-1.7B/checkpoint-512",
|
|
"epoch": 0.22394751230180426,
|
|
"eval_steps": 256,
|
|
"global_step": 512,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 0.27786555141210556,
|
|
"epoch": 0.00043739748496446145,
|
|
"grad_norm": 21.25,
|
|
"learning_rate": 0.0,
|
|
"loss": 0.901656985282898,
|
|
"mean_token_accuracy": 0.8019306063652039,
|
|
"num_tokens": 8192.0,
|
|
"step": 1
|
|
},
|
|
{
|
|
"entropy": 0.32270805910229683,
|
|
"epoch": 0.0008747949699289229,
|
|
"grad_norm": 20.75,
|
|
"learning_rate": 7.8125e-08,
|
|
"loss": 0.9285624027252197,
|
|
"mean_token_accuracy": 0.7913001030683517,
|
|
"num_tokens": 16384.0,
|
|
"step": 2
|
|
},
|
|
{
|
|
"entropy": 0.3030129224061966,
|
|
"epoch": 0.0013121924548933845,
|
|
"grad_norm": 23.375,
|
|
"learning_rate": 1.5625e-07,
|
|
"loss": 0.9610694646835327,
|
|
"mean_token_accuracy": 0.7899560034275055,
|
|
"num_tokens": 24576.0,
|
|
"step": 3
|
|
},
|
|
{
|
|
"entropy": 0.3043339401483536,
|
|
"epoch": 0.0017495899398578458,
|
|
"grad_norm": 22.0,
|
|
"learning_rate": 2.3437500000000003e-07,
|
|
"loss": 0.9333600997924805,
|
|
"mean_token_accuracy": 0.7908113300800323,
|
|
"num_tokens": 32768.0,
|
|
"step": 4
|
|
},
|
|
{
|
|
"entropy": 0.30834493041038513,
|
|
"epoch": 0.002186987424822307,
|
|
"grad_norm": 21.75,
|
|
"learning_rate": 3.125e-07,
|
|
"loss": 0.9301238059997559,
|
|
"mean_token_accuracy": 0.7915444821119308,
|
|
"num_tokens": 40960.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 0.3122338131070137,
|
|
"epoch": 0.002624384909786769,
|
|
"grad_norm": 22.5,
|
|
"learning_rate": 3.90625e-07,
|
|
"loss": 0.9689892530441284,
|
|
"mean_token_accuracy": 0.7847018390893936,
|
|
"num_tokens": 49152.0,
|
|
"step": 6
|
|
},
|
|
{
|
|
"entropy": 0.2893872708082199,
|
|
"epoch": 0.0030617823947512303,
|
|
"grad_norm": 21.0,
|
|
"learning_rate": 4.6875000000000006e-07,
|
|
"loss": 0.9168861508369446,
|
|
"mean_token_accuracy": 0.8013196438550949,
|
|
"num_tokens": 57344.0,
|
|
"step": 7
|
|
},
|
|
{
|
|
"entropy": 0.3050616979598999,
|
|
"epoch": 0.0034991798797156916,
|
|
"grad_norm": 21.875,
|
|
"learning_rate": 5.468750000000001e-07,
|
|
"loss": 0.9516887664794922,
|
|
"mean_token_accuracy": 0.7850684374570847,
|
|
"num_tokens": 65536.0,
|
|
"step": 8
|
|
},
|
|
{
|
|
"entropy": 0.31983715295791626,
|
|
"epoch": 0.003936577364680153,
|
|
"grad_norm": 23.375,
|
|
"learning_rate": 6.25e-07,
|
|
"loss": 0.9918694496154785,
|
|
"mean_token_accuracy": 0.7829912006855011,
|
|
"num_tokens": 73728.0,
|
|
"step": 9
|
|
},
|
|
{
|
|
"entropy": 0.32186224311590195,
|
|
"epoch": 0.004373974849644614,
|
|
"grad_norm": 21.0,
|
|
"learning_rate": 7.03125e-07,
|
|
"loss": 0.9560595750808716,
|
|
"mean_token_accuracy": 0.7919110506772995,
|
|
"num_tokens": 81920.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 0.32521846890449524,
|
|
"epoch": 0.004811372334609076,
|
|
"grad_norm": 22.875,
|
|
"learning_rate": 7.8125e-07,
|
|
"loss": 0.9732818007469177,
|
|
"mean_token_accuracy": 0.7860459387302399,
|
|
"num_tokens": 90112.0,
|
|
"step": 11
|
|
},
|
|
{
|
|
"entropy": 0.31955306977033615,
|
|
"epoch": 0.005248769819573538,
|
|
"grad_norm": 21.75,
|
|
"learning_rate": 8.59375e-07,
|
|
"loss": 0.9713459014892578,
|
|
"mean_token_accuracy": 0.7805474102497101,
|
|
"num_tokens": 98304.0,
|
|
"step": 12
|
|
},
|
|
{
|
|
"entropy": 0.30780235305428505,
|
|
"epoch": 0.005686167304537999,
|
|
"grad_norm": 20.625,
|
|
"learning_rate": 9.375000000000001e-07,
|
|
"loss": 0.9195499420166016,
|
|
"mean_token_accuracy": 0.7969208359718323,
|
|
"num_tokens": 106496.0,
|
|
"step": 13
|
|
},
|
|
{
|
|
"entropy": 0.33620841801166534,
|
|
"epoch": 0.0061235647895024605,
|
|
"grad_norm": 21.25,
|
|
"learning_rate": 1.0156250000000001e-06,
|
|
"loss": 0.9759924411773682,
|
|
"mean_token_accuracy": 0.7779259830713272,
|
|
"num_tokens": 114652.0,
|
|
"step": 14
|
|
},
|
|
{
|
|
"entropy": 0.30951154232025146,
|
|
"epoch": 0.0065609622744669215,
|
|
"grad_norm": 20.0,
|
|
"learning_rate": 1.0937500000000001e-06,
|
|
"loss": 0.8884122967720032,
|
|
"mean_token_accuracy": 0.7992424219846725,
|
|
"num_tokens": 122844.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 0.3126643896102905,
|
|
"epoch": 0.006998359759431383,
|
|
"grad_norm": 20.5,
|
|
"learning_rate": 1.1718750000000001e-06,
|
|
"loss": 0.9235544204711914,
|
|
"mean_token_accuracy": 0.7932551354169846,
|
|
"num_tokens": 131036.0,
|
|
"step": 16
|
|
},
|
|
{
|
|
"entropy": 0.32217270135879517,
|
|
"epoch": 0.007435757244395845,
|
|
"grad_norm": 22.25,
|
|
"learning_rate": 1.25e-06,
|
|
"loss": 0.9568185210227966,
|
|
"mean_token_accuracy": 0.7825024574995041,
|
|
"num_tokens": 139228.0,
|
|
"step": 17
|
|
},
|
|
{
|
|
"entropy": 0.28684747219085693,
|
|
"epoch": 0.007873154729360307,
|
|
"grad_norm": 18.25,
|
|
"learning_rate": 1.328125e-06,
|
|
"loss": 0.8255479335784912,
|
|
"mean_token_accuracy": 0.8137829899787903,
|
|
"num_tokens": 147420.0,
|
|
"step": 18
|
|
},
|
|
{
|
|
"entropy": 0.3058253973722458,
|
|
"epoch": 0.008310552214324767,
|
|
"grad_norm": 20.0,
|
|
"learning_rate": 1.40625e-06,
|
|
"loss": 0.8979604244232178,
|
|
"mean_token_accuracy": 0.7972874045372009,
|
|
"num_tokens": 155612.0,
|
|
"step": 19
|
|
},
|
|
{
|
|
"entropy": 0.38854775577783585,
|
|
"epoch": 0.008747949699289229,
|
|
"grad_norm": 22.625,
|
|
"learning_rate": 1.484375e-06,
|
|
"loss": 1.0705394744873047,
|
|
"mean_token_accuracy": 0.7572091966867447,
|
|
"num_tokens": 163804.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 0.30834781378507614,
|
|
"epoch": 0.00918534718425369,
|
|
"grad_norm": 20.375,
|
|
"learning_rate": 1.5625e-06,
|
|
"loss": 0.9028609395027161,
|
|
"mean_token_accuracy": 0.7923997938632965,
|
|
"num_tokens": 171996.0,
|
|
"step": 21
|
|
},
|
|
{
|
|
"entropy": 0.32601282000541687,
|
|
"epoch": 0.009622744669218152,
|
|
"grad_norm": 20.625,
|
|
"learning_rate": 1.640625e-06,
|
|
"loss": 0.9245335459709167,
|
|
"mean_token_accuracy": 0.7817693054676056,
|
|
"num_tokens": 180188.0,
|
|
"step": 22
|
|
},
|
|
{
|
|
"entropy": 0.30635514110326767,
|
|
"epoch": 0.010060142154182614,
|
|
"grad_norm": 19.875,
|
|
"learning_rate": 1.71875e-06,
|
|
"loss": 0.9333948493003845,
|
|
"mean_token_accuracy": 0.7876344174146652,
|
|
"num_tokens": 188380.0,
|
|
"step": 23
|
|
},
|
|
{
|
|
"entropy": 0.3130311816930771,
|
|
"epoch": 0.010497539639147076,
|
|
"grad_norm": 19.625,
|
|
"learning_rate": 1.796875e-06,
|
|
"loss": 0.8786435723304749,
|
|
"mean_token_accuracy": 0.801075279712677,
|
|
"num_tokens": 196572.0,
|
|
"step": 24
|
|
},
|
|
{
|
|
"entropy": 0.3683779239654541,
|
|
"epoch": 0.010934937124111536,
|
|
"grad_norm": 21.0,
|
|
"learning_rate": 1.8750000000000003e-06,
|
|
"loss": 0.9972445964813232,
|
|
"mean_token_accuracy": 0.7724829018115997,
|
|
"num_tokens": 204764.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 0.34924668073654175,
|
|
"epoch": 0.011372334609075998,
|
|
"grad_norm": 22.125,
|
|
"learning_rate": 1.953125e-06,
|
|
"loss": 1.0172064304351807,
|
|
"mean_token_accuracy": 0.7709844559431076,
|
|
"num_tokens": 212677.0,
|
|
"step": 26
|
|
},
|
|
{
|
|
"entropy": 0.3614501357078552,
|
|
"epoch": 0.01180973209404046,
|
|
"grad_norm": 19.625,
|
|
"learning_rate": 2.0312500000000002e-06,
|
|
"loss": 0.9596425294876099,
|
|
"mean_token_accuracy": 0.7776148617267609,
|
|
"num_tokens": 220869.0,
|
|
"step": 27
|
|
},
|
|
{
|
|
"entropy": 0.31008924543857574,
|
|
"epoch": 0.012247129579004921,
|
|
"grad_norm": 18.25,
|
|
"learning_rate": 2.109375e-06,
|
|
"loss": 0.8647151589393616,
|
|
"mean_token_accuracy": 0.7933773398399353,
|
|
"num_tokens": 229061.0,
|
|
"step": 28
|
|
},
|
|
{
|
|
"entropy": 0.29904749244451523,
|
|
"epoch": 0.012684527063969383,
|
|
"grad_norm": 18.875,
|
|
"learning_rate": 2.1875000000000002e-06,
|
|
"loss": 0.8574439287185669,
|
|
"mean_token_accuracy": 0.804007813334465,
|
|
"num_tokens": 237253.0,
|
|
"step": 29
|
|
},
|
|
{
|
|
"entropy": 0.378295473754406,
|
|
"epoch": 0.013121924548933843,
|
|
"grad_norm": 19.75,
|
|
"learning_rate": 2.265625e-06,
|
|
"loss": 0.9890425205230713,
|
|
"mean_token_accuracy": 0.7633186727762222,
|
|
"num_tokens": 245445.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 0.360466867685318,
|
|
"epoch": 0.013559322033898305,
|
|
"grad_norm": 18.0,
|
|
"learning_rate": 2.3437500000000002e-06,
|
|
"loss": 0.9291175603866577,
|
|
"mean_token_accuracy": 0.7820213884115219,
|
|
"num_tokens": 253493.0,
|
|
"step": 31
|
|
},
|
|
{
|
|
"entropy": 0.30008650943636894,
|
|
"epoch": 0.013996719518862766,
|
|
"grad_norm": 16.875,
|
|
"learning_rate": 2.421875e-06,
|
|
"loss": 0.7958075404167175,
|
|
"mean_token_accuracy": 0.8091397881507874,
|
|
"num_tokens": 261685.0,
|
|
"step": 32
|
|
},
|
|
{
|
|
"entropy": 0.3233437053859234,
|
|
"epoch": 0.014434117003827228,
|
|
"grad_norm": 16.75,
|
|
"learning_rate": 2.5e-06,
|
|
"loss": 0.8411202430725098,
|
|
"mean_token_accuracy": 0.793621689081192,
|
|
"num_tokens": 269877.0,
|
|
"step": 33
|
|
},
|
|
{
|
|
"entropy": 0.30822817981243134,
|
|
"epoch": 0.01487151448879169,
|
|
"grad_norm": 16.75,
|
|
"learning_rate": 2.5781250000000004e-06,
|
|
"loss": 0.8265758752822876,
|
|
"mean_token_accuracy": 0.8031525015830994,
|
|
"num_tokens": 278069.0,
|
|
"step": 34
|
|
},
|
|
{
|
|
"entropy": 0.3806365579366684,
|
|
"epoch": 0.01530891197375615,
|
|
"grad_norm": 15.4375,
|
|
"learning_rate": 2.65625e-06,
|
|
"loss": 0.8731443881988525,
|
|
"mean_token_accuracy": 0.7812805324792862,
|
|
"num_tokens": 286261.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 0.28861086070537567,
|
|
"epoch": 0.015746309458720614,
|
|
"grad_norm": 14.0625,
|
|
"learning_rate": 2.7343750000000004e-06,
|
|
"loss": 0.7268953323364258,
|
|
"mean_token_accuracy": 0.822182759642601,
|
|
"num_tokens": 294432.0,
|
|
"step": 36
|
|
},
|
|
{
|
|
"entropy": 0.333795465528965,
|
|
"epoch": 0.016183706943685074,
|
|
"grad_norm": 15.875,
|
|
"learning_rate": 2.8125e-06,
|
|
"loss": 0.8444324731826782,
|
|
"mean_token_accuracy": 0.7978983372449875,
|
|
"num_tokens": 302624.0,
|
|
"step": 37
|
|
},
|
|
{
|
|
"entropy": 0.33112896978855133,
|
|
"epoch": 0.016621104428649534,
|
|
"grad_norm": 15.75,
|
|
"learning_rate": 2.8906250000000004e-06,
|
|
"loss": 0.8359007835388184,
|
|
"mean_token_accuracy": 0.792033240199089,
|
|
"num_tokens": 310816.0,
|
|
"step": 38
|
|
},
|
|
{
|
|
"entropy": 0.34320055693387985,
|
|
"epoch": 0.017058501913613997,
|
|
"grad_norm": 13.0,
|
|
"learning_rate": 2.96875e-06,
|
|
"loss": 0.7701590061187744,
|
|
"mean_token_accuracy": 0.8037634491920471,
|
|
"num_tokens": 319008.0,
|
|
"step": 39
|
|
},
|
|
{
|
|
"entropy": 0.42401332408189774,
|
|
"epoch": 0.017495899398578457,
|
|
"grad_norm": 14.3125,
|
|
"learning_rate": 3.0468750000000004e-06,
|
|
"loss": 0.8930312395095825,
|
|
"mean_token_accuracy": 0.7701613008975983,
|
|
"num_tokens": 327200.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 0.3282608240842819,
|
|
"epoch": 0.01793329688354292,
|
|
"grad_norm": 12.875,
|
|
"learning_rate": 3.125e-06,
|
|
"loss": 0.7595839500427246,
|
|
"mean_token_accuracy": 0.8081848919391632,
|
|
"num_tokens": 335128.0,
|
|
"step": 41
|
|
},
|
|
{
|
|
"entropy": 0.3341097980737686,
|
|
"epoch": 0.01837069436850738,
|
|
"grad_norm": 12.6875,
|
|
"learning_rate": 3.2031250000000004e-06,
|
|
"loss": 0.7940071821212769,
|
|
"mean_token_accuracy": 0.8011974543333054,
|
|
"num_tokens": 343320.0,
|
|
"step": 42
|
|
},
|
|
{
|
|
"entropy": 0.35623009502887726,
|
|
"epoch": 0.01880809185347184,
|
|
"grad_norm": 13.5625,
|
|
"learning_rate": 3.28125e-06,
|
|
"loss": 0.7961429357528687,
|
|
"mean_token_accuracy": 0.7959433048963547,
|
|
"num_tokens": 351512.0,
|
|
"step": 43
|
|
},
|
|
{
|
|
"entropy": 0.33236806094646454,
|
|
"epoch": 0.019245489338436304,
|
|
"grad_norm": 12.8125,
|
|
"learning_rate": 3.3593750000000003e-06,
|
|
"loss": 0.7723078727722168,
|
|
"mean_token_accuracy": 0.8073069304227829,
|
|
"num_tokens": 359704.0,
|
|
"step": 44
|
|
},
|
|
{
|
|
"entropy": 0.45774705708026886,
|
|
"epoch": 0.019682886823400764,
|
|
"grad_norm": 13.9375,
|
|
"learning_rate": 3.4375e-06,
|
|
"loss": 0.9575591087341309,
|
|
"mean_token_accuracy": 0.757942333817482,
|
|
"num_tokens": 367896.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 0.35400258004665375,
|
|
"epoch": 0.020120284308365228,
|
|
"grad_norm": 10.375,
|
|
"learning_rate": 3.5156250000000003e-06,
|
|
"loss": 0.719309389591217,
|
|
"mean_token_accuracy": 0.814393937587738,
|
|
"num_tokens": 376088.0,
|
|
"step": 46
|
|
},
|
|
{
|
|
"entropy": 0.3176649585366249,
|
|
"epoch": 0.020557681793329688,
|
|
"grad_norm": 9.4375,
|
|
"learning_rate": 3.59375e-06,
|
|
"loss": 0.6449992656707764,
|
|
"mean_token_accuracy": 0.8264907151460648,
|
|
"num_tokens": 384280.0,
|
|
"step": 47
|
|
},
|
|
{
|
|
"entropy": 0.3971637114882469,
|
|
"epoch": 0.02099507927829415,
|
|
"grad_norm": 11.4375,
|
|
"learning_rate": 3.6718750000000003e-06,
|
|
"loss": 0.8187000155448914,
|
|
"mean_token_accuracy": 0.7884897291660309,
|
|
"num_tokens": 392472.0,
|
|
"step": 48
|
|
},
|
|
{
|
|
"entropy": 0.42229485511779785,
|
|
"epoch": 0.02143247676325861,
|
|
"grad_norm": 11.6875,
|
|
"learning_rate": 3.7500000000000005e-06,
|
|
"loss": 0.846123456954956,
|
|
"mean_token_accuracy": 0.7745601385831833,
|
|
"num_tokens": 400664.0,
|
|
"step": 49
|
|
},
|
|
{
|
|
"entropy": 0.3986642137169838,
|
|
"epoch": 0.02186987424822307,
|
|
"grad_norm": 10.625,
|
|
"learning_rate": 3.828125000000001e-06,
|
|
"loss": 0.7735276222229004,
|
|
"mean_token_accuracy": 0.7872678488492966,
|
|
"num_tokens": 408856.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 0.43453069031238556,
|
|
"epoch": 0.022307271733187535,
|
|
"grad_norm": 11.5625,
|
|
"learning_rate": 3.90625e-06,
|
|
"loss": 0.8254950046539307,
|
|
"mean_token_accuracy": 0.7809139937162399,
|
|
"num_tokens": 417048.0,
|
|
"step": 51
|
|
},
|
|
{
|
|
"entropy": 0.37958449125289917,
|
|
"epoch": 0.022744669218151995,
|
|
"grad_norm": 8.75,
|
|
"learning_rate": 3.984375e-06,
|
|
"loss": 0.6892186403274536,
|
|
"mean_token_accuracy": 0.813049852848053,
|
|
"num_tokens": 425240.0,
|
|
"step": 52
|
|
},
|
|
{
|
|
"entropy": 0.4048294723033905,
|
|
"epoch": 0.02318206670311646,
|
|
"grad_norm": 8.8125,
|
|
"learning_rate": 4.0625000000000005e-06,
|
|
"loss": 0.7329984307289124,
|
|
"mean_token_accuracy": 0.798020526766777,
|
|
"num_tokens": 433432.0,
|
|
"step": 53
|
|
},
|
|
{
|
|
"entropy": 0.42267417162656784,
|
|
"epoch": 0.02361946418808092,
|
|
"grad_norm": 9.1875,
|
|
"learning_rate": 4.140625000000001e-06,
|
|
"loss": 0.7617641687393188,
|
|
"mean_token_accuracy": 0.7957284301519394,
|
|
"num_tokens": 441608.0,
|
|
"step": 54
|
|
},
|
|
{
|
|
"entropy": 0.44952575117349625,
|
|
"epoch": 0.02405686167304538,
|
|
"grad_norm": 8.125,
|
|
"learning_rate": 4.21875e-06,
|
|
"loss": 0.7307311296463013,
|
|
"mean_token_accuracy": 0.787634402513504,
|
|
"num_tokens": 449800.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 0.45996539294719696,
|
|
"epoch": 0.024494259158009842,
|
|
"grad_norm": 7.4375,
|
|
"learning_rate": 4.296875e-06,
|
|
"loss": 0.7325446605682373,
|
|
"mean_token_accuracy": 0.7894672453403473,
|
|
"num_tokens": 457992.0,
|
|
"step": 56
|
|
},
|
|
{
|
|
"entropy": 0.46281183511018753,
|
|
"epoch": 0.024931656642974302,
|
|
"grad_norm": 7.78125,
|
|
"learning_rate": 4.3750000000000005e-06,
|
|
"loss": 0.7531175017356873,
|
|
"mean_token_accuracy": 0.7916666567325592,
|
|
"num_tokens": 466184.0,
|
|
"step": 57
|
|
},
|
|
{
|
|
"entropy": 0.40494948625564575,
|
|
"epoch": 0.025369054127938766,
|
|
"grad_norm": 6.46875,
|
|
"learning_rate": 4.453125000000001e-06,
|
|
"loss": 0.6567135453224182,
|
|
"mean_token_accuracy": 0.8132942169904709,
|
|
"num_tokens": 474376.0,
|
|
"step": 58
|
|
},
|
|
{
|
|
"entropy": 0.44023270159959793,
|
|
"epoch": 0.025806451612903226,
|
|
"grad_norm": 7.25,
|
|
"learning_rate": 4.53125e-06,
|
|
"loss": 0.703899621963501,
|
|
"mean_token_accuracy": 0.8019305914640427,
|
|
"num_tokens": 482568.0,
|
|
"step": 59
|
|
},
|
|
{
|
|
"entropy": 0.3879058510065079,
|
|
"epoch": 0.026243849097867686,
|
|
"grad_norm": 6.15625,
|
|
"learning_rate": 4.609375e-06,
|
|
"loss": 0.6011914014816284,
|
|
"mean_token_accuracy": 0.8233137875795364,
|
|
"num_tokens": 490760.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 0.41236213594675064,
|
|
"epoch": 0.02668124658283215,
|
|
"grad_norm": 5.46875,
|
|
"learning_rate": 4.6875000000000004e-06,
|
|
"loss": 0.5771687626838684,
|
|
"mean_token_accuracy": 0.8264907002449036,
|
|
"num_tokens": 498952.0,
|
|
"step": 61
|
|
},
|
|
{
|
|
"entropy": 0.4559241756796837,
|
|
"epoch": 0.02711864406779661,
|
|
"grad_norm": 5.59375,
|
|
"learning_rate": 4.765625000000001e-06,
|
|
"loss": 0.6721391081809998,
|
|
"mean_token_accuracy": 0.8046199232339859,
|
|
"num_tokens": 507024.0,
|
|
"step": 62
|
|
},
|
|
{
|
|
"entropy": 0.40718207508325577,
|
|
"epoch": 0.027556041552761073,
|
|
"grad_norm": 5.15625,
|
|
"learning_rate": 4.84375e-06,
|
|
"loss": 0.5817515850067139,
|
|
"mean_token_accuracy": 0.8244134932756424,
|
|
"num_tokens": 515216.0,
|
|
"step": 63
|
|
},
|
|
{
|
|
"entropy": 0.4880864918231964,
|
|
"epoch": 0.027993439037725533,
|
|
"grad_norm": 5.1875,
|
|
"learning_rate": 4.921875e-06,
|
|
"loss": 0.6458011865615845,
|
|
"mean_token_accuracy": 0.8033017814159393,
|
|
"num_tokens": 523359.0,
|
|
"step": 64
|
|
},
|
|
{
|
|
"entropy": 0.4802205115556717,
|
|
"epoch": 0.028430836522689993,
|
|
"grad_norm": 4.8125,
|
|
"learning_rate": 5e-06,
|
|
"loss": 0.6212886571884155,
|
|
"mean_token_accuracy": 0.8052297234535217,
|
|
"num_tokens": 531551.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 0.5558610707521439,
|
|
"epoch": 0.028868234007654456,
|
|
"grad_norm": 5.6875,
|
|
"learning_rate": 4.999938531594062e-06,
|
|
"loss": 0.7324199676513672,
|
|
"mean_token_accuracy": 0.7729716449975967,
|
|
"num_tokens": 539743.0,
|
|
"step": 66
|
|
},
|
|
{
|
|
"entropy": 0.4111747443675995,
|
|
"epoch": 0.029305631492618917,
|
|
"grad_norm": 4.0,
|
|
"learning_rate": 4.999754129398938e-06,
|
|
"loss": 0.5279664993286133,
|
|
"mean_token_accuracy": 0.8387096673250198,
|
|
"num_tokens": 547935.0,
|
|
"step": 67
|
|
},
|
|
{
|
|
"entropy": 0.49969246983528137,
|
|
"epoch": 0.02974302897758338,
|
|
"grad_norm": 4.21875,
|
|
"learning_rate": 4.999446802482556e-06,
|
|
"loss": 0.6269458532333374,
|
|
"mean_token_accuracy": 0.807062566280365,
|
|
"num_tokens": 556127.0,
|
|
"step": 68
|
|
},
|
|
{
|
|
"entropy": 0.533087246119976,
|
|
"epoch": 0.03018042646254784,
|
|
"grad_norm": 4.46875,
|
|
"learning_rate": 4.999016565957633e-06,
|
|
"loss": 0.6664482355117798,
|
|
"mean_token_accuracy": 0.7987536638975143,
|
|
"num_tokens": 564319.0,
|
|
"step": 69
|
|
},
|
|
{
|
|
"entropy": 0.4802093654870987,
|
|
"epoch": 0.0306178239475123,
|
|
"grad_norm": 3.984375,
|
|
"learning_rate": 4.998463440980931e-06,
|
|
"loss": 0.5956555604934692,
|
|
"mean_token_accuracy": 0.8218475133180618,
|
|
"num_tokens": 572511.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 0.5158685892820358,
|
|
"epoch": 0.031055221432476764,
|
|
"grad_norm": 4.3125,
|
|
"learning_rate": 4.997787454752217e-06,
|
|
"loss": 0.6313604116439819,
|
|
"mean_token_accuracy": 0.8016862124204636,
|
|
"num_tokens": 580703.0,
|
|
"step": 71
|
|
},
|
|
{
|
|
"entropy": 0.49100108444690704,
|
|
"epoch": 0.03149261891744123,
|
|
"grad_norm": 3.8125,
|
|
"learning_rate": 4.996988640512931e-06,
|
|
"loss": 0.595371663570404,
|
|
"mean_token_accuracy": 0.8114613741636276,
|
|
"num_tokens": 588895.0,
|
|
"step": 72
|
|
},
|
|
{
|
|
"entropy": 0.5480513125658035,
|
|
"epoch": 0.031930016402405684,
|
|
"grad_norm": 3.90625,
|
|
"learning_rate": 4.996067037544542e-06,
|
|
"loss": 0.643286406993866,
|
|
"mean_token_accuracy": 0.7959433048963547,
|
|
"num_tokens": 597087.0,
|
|
"step": 73
|
|
},
|
|
{
|
|
"entropy": 0.5591509789228439,
|
|
"epoch": 0.03236741388737015,
|
|
"grad_norm": 4.625,
|
|
"learning_rate": 4.995022691166621e-06,
|
|
"loss": 0.6653096675872803,
|
|
"mean_token_accuracy": 0.7919110357761383,
|
|
"num_tokens": 605279.0,
|
|
"step": 74
|
|
},
|
|
{
|
|
"entropy": 0.5032752007246017,
|
|
"epoch": 0.03280481137233461,
|
|
"grad_norm": 3.640625,
|
|
"learning_rate": 4.993855652734616e-06,
|
|
"loss": 0.5787771940231323,
|
|
"mean_token_accuracy": 0.8167155385017395,
|
|
"num_tokens": 613471.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 0.4510410279035568,
|
|
"epoch": 0.03324220885729907,
|
|
"grad_norm": 3.453125,
|
|
"learning_rate": 4.992565979637318e-06,
|
|
"loss": 0.5281908512115479,
|
|
"mean_token_accuracy": 0.8304007798433304,
|
|
"num_tokens": 621663.0,
|
|
"step": 76
|
|
},
|
|
{
|
|
"entropy": 0.4650597348809242,
|
|
"epoch": 0.03367960634226353,
|
|
"grad_norm": 3.640625,
|
|
"learning_rate": 4.9911537352940485e-06,
|
|
"loss": 0.5683005452156067,
|
|
"mean_token_accuracy": 0.8246578723192215,
|
|
"num_tokens": 629855.0,
|
|
"step": 77
|
|
},
|
|
{
|
|
"entropy": 0.5157822743058205,
|
|
"epoch": 0.034117003827227994,
|
|
"grad_norm": 3.9375,
|
|
"learning_rate": 4.989618989151533e-06,
|
|
"loss": 0.5880601406097412,
|
|
"mean_token_accuracy": 0.8147605061531067,
|
|
"num_tokens": 638047.0,
|
|
"step": 78
|
|
},
|
|
{
|
|
"entropy": 0.58241768181324,
|
|
"epoch": 0.03455440131219246,
|
|
"grad_norm": 3.859375,
|
|
"learning_rate": 4.987961816680493e-06,
|
|
"loss": 0.6643019914627075,
|
|
"mean_token_accuracy": 0.794232651591301,
|
|
"num_tokens": 646239.0,
|
|
"step": 79
|
|
},
|
|
{
|
|
"entropy": 0.46477261185646057,
|
|
"epoch": 0.034991798797156914,
|
|
"grad_norm": 3.03125,
|
|
"learning_rate": 4.9861822993719255e-06,
|
|
"loss": 0.5226079225540161,
|
|
"mean_token_accuracy": 0.8370988517999649,
|
|
"num_tokens": 654199.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 0.4996022880077362,
|
|
"epoch": 0.03542919628212138,
|
|
"grad_norm": 3.4375,
|
|
"learning_rate": 4.984280524733107e-06,
|
|
"loss": 0.5533886551856995,
|
|
"mean_token_accuracy": 0.8238108903169632,
|
|
"num_tokens": 662335.0,
|
|
"step": 81
|
|
},
|
|
{
|
|
"entropy": 0.5934857875108719,
|
|
"epoch": 0.03586659376708584,
|
|
"grad_norm": 3.6875,
|
|
"learning_rate": 4.98225658628328e-06,
|
|
"loss": 0.6762368679046631,
|
|
"mean_token_accuracy": 0.7827468067407608,
|
|
"num_tokens": 670527.0,
|
|
"step": 82
|
|
},
|
|
{
|
|
"entropy": 0.563569039106369,
|
|
"epoch": 0.0363039912520503,
|
|
"grad_norm": 3.390625,
|
|
"learning_rate": 4.980110583549062e-06,
|
|
"loss": 0.6458903551101685,
|
|
"mean_token_accuracy": 0.7992424219846725,
|
|
"num_tokens": 678719.0,
|
|
"step": 83
|
|
},
|
|
{
|
|
"entropy": 0.4253915399312973,
|
|
"epoch": 0.03674138873701476,
|
|
"grad_norm": 2.6875,
|
|
"learning_rate": 4.977842622059546e-06,
|
|
"loss": 0.47188764810562134,
|
|
"mean_token_accuracy": 0.8451857268810272,
|
|
"num_tokens": 686911.0,
|
|
"step": 84
|
|
},
|
|
{
|
|
"entropy": 0.5138464272022247,
|
|
"epoch": 0.037178786221979225,
|
|
"grad_norm": 3.140625,
|
|
"learning_rate": 4.9754528133411144e-06,
|
|
"loss": 0.5898129940032959,
|
|
"mean_token_accuracy": 0.8066959828138351,
|
|
"num_tokens": 695103.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 0.5432897508144379,
|
|
"epoch": 0.03761618370694368,
|
|
"grad_norm": 3.140625,
|
|
"learning_rate": 4.972941274911953e-06,
|
|
"loss": 0.6102941632270813,
|
|
"mean_token_accuracy": 0.8013196438550949,
|
|
"num_tokens": 703295.0,
|
|
"step": 86
|
|
},
|
|
{
|
|
"entropy": 0.4696696624159813,
|
|
"epoch": 0.038053581191908145,
|
|
"grad_norm": 3.109375,
|
|
"learning_rate": 4.970308130276273e-06,
|
|
"loss": 0.5391623973846436,
|
|
"mean_token_accuracy": 0.8216031193733215,
|
|
"num_tokens": 711487.0,
|
|
"step": 87
|
|
},
|
|
{
|
|
"entropy": 0.45806846022605896,
|
|
"epoch": 0.03849097867687261,
|
|
"grad_norm": 2.90625,
|
|
"learning_rate": 4.967553508918236e-06,
|
|
"loss": 0.5263965129852295,
|
|
"mean_token_accuracy": 0.8305229693651199,
|
|
"num_tokens": 719679.0,
|
|
"step": 88
|
|
},
|
|
{
|
|
"entropy": 0.49067409336566925,
|
|
"epoch": 0.03892837616183707,
|
|
"grad_norm": 3.0,
|
|
"learning_rate": 4.96467754629559e-06,
|
|
"loss": 0.5333690047264099,
|
|
"mean_token_accuracy": 0.8266129046678543,
|
|
"num_tokens": 727871.0,
|
|
"step": 89
|
|
},
|
|
{
|
|
"entropy": 0.5586400255560875,
|
|
"epoch": 0.03936577364680153,
|
|
"grad_norm": 3.296875,
|
|
"learning_rate": 4.961680383833005e-06,
|
|
"loss": 0.6310930848121643,
|
|
"mean_token_accuracy": 0.7937438935041428,
|
|
"num_tokens": 736063.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 0.6168518960475922,
|
|
"epoch": 0.03980317113176599,
|
|
"grad_norm": 3.296875,
|
|
"learning_rate": 4.958562168915121e-06,
|
|
"loss": 0.7125385999679565,
|
|
"mean_token_accuracy": 0.7805473953485489,
|
|
"num_tokens": 744255.0,
|
|
"step": 91
|
|
},
|
|
{
|
|
"entropy": 0.5408408343791962,
|
|
"epoch": 0.040240568616730456,
|
|
"grad_norm": 3.0,
|
|
"learning_rate": 4.955323054879298e-06,
|
|
"loss": 0.6216639280319214,
|
|
"mean_token_accuracy": 0.8088954240083694,
|
|
"num_tokens": 752447.0,
|
|
"step": 92
|
|
},
|
|
{
|
|
"entropy": 0.445939376950264,
|
|
"epoch": 0.04067796610169491,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 4.9519632010080765e-06,
|
|
"loss": 0.4946935772895813,
|
|
"mean_token_accuracy": 0.8440860360860825,
|
|
"num_tokens": 760639.0,
|
|
"step": 93
|
|
},
|
|
{
|
|
"entropy": 0.346243679523468,
|
|
"epoch": 0.041115363586659376,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 4.948482772521346e-06,
|
|
"loss": 0.38514432311058044,
|
|
"mean_token_accuracy": 0.8690126985311508,
|
|
"num_tokens": 768831.0,
|
|
"step": 94
|
|
},
|
|
{
|
|
"entropy": 0.5287810266017914,
|
|
"epoch": 0.04155276107162384,
|
|
"grad_norm": 2.90625,
|
|
"learning_rate": 4.944881940568219e-06,
|
|
"loss": 0.5804294943809509,
|
|
"mean_token_accuracy": 0.8090175986289978,
|
|
"num_tokens": 777023.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 0.4812530502676964,
|
|
"epoch": 0.0419901585565883,
|
|
"grad_norm": 2.75,
|
|
"learning_rate": 4.941160882218617e-06,
|
|
"loss": 0.5441983938217163,
|
|
"mean_token_accuracy": 0.8217253088951111,
|
|
"num_tokens": 785215.0,
|
|
"step": 96
|
|
},
|
|
{
|
|
"entropy": 0.5249937623739243,
|
|
"epoch": 0.04242755604155276,
|
|
"grad_norm": 3.171875,
|
|
"learning_rate": 4.937319780454559e-06,
|
|
"loss": 0.6077675819396973,
|
|
"mean_token_accuracy": 0.8069403767585754,
|
|
"num_tokens": 793407.0,
|
|
"step": 97
|
|
},
|
|
{
|
|
"entropy": 0.5760547369718552,
|
|
"epoch": 0.04286495352651722,
|
|
"grad_norm": 3.0,
|
|
"learning_rate": 4.933358824161167e-06,
|
|
"loss": 0.6388422250747681,
|
|
"mean_token_accuracy": 0.7965542525053024,
|
|
"num_tokens": 801599.0,
|
|
"step": 98
|
|
},
|
|
{
|
|
"entropy": 0.4914955049753189,
|
|
"epoch": 0.043302351011481686,
|
|
"grad_norm": 2.734375,
|
|
"learning_rate": 4.929278208117378e-06,
|
|
"loss": 0.5518084168434143,
|
|
"mean_token_accuracy": 0.8212365508079529,
|
|
"num_tokens": 809791.0,
|
|
"step": 99
|
|
},
|
|
{
|
|
"entropy": 0.4663372188806534,
|
|
"epoch": 0.04373974849644614,
|
|
"grad_norm": 2.703125,
|
|
"learning_rate": 4.925078132986361e-06,
|
|
"loss": 0.521884560585022,
|
|
"mean_token_accuracy": 0.8260019570589066,
|
|
"num_tokens": 817983.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 0.5335247218608856,
|
|
"epoch": 0.04417714598141061,
|
|
"grad_norm": 2.859375,
|
|
"learning_rate": 4.920758805305654e-06,
|
|
"loss": 0.5998464822769165,
|
|
"mean_token_accuracy": 0.8064516186714172,
|
|
"num_tokens": 826175.0,
|
|
"step": 101
|
|
},
|
|
{
|
|
"entropy": 0.5472059100866318,
|
|
"epoch": 0.04461454346637507,
|
|
"grad_norm": 2.953125,
|
|
"learning_rate": 4.9163204374770085e-06,
|
|
"loss": 0.6169013977050781,
|
|
"mean_token_accuracy": 0.8025415539741516,
|
|
"num_tokens": 834367.0,
|
|
"step": 102
|
|
},
|
|
{
|
|
"entropy": 0.615875169634819,
|
|
"epoch": 0.04505194095133953,
|
|
"grad_norm": 2.9375,
|
|
"learning_rate": 4.91176324775594e-06,
|
|
"loss": 0.6760657429695129,
|
|
"mean_token_accuracy": 0.7895894348621368,
|
|
"num_tokens": 842559.0,
|
|
"step": 103
|
|
},
|
|
{
|
|
"entropy": 0.36748453229665756,
|
|
"epoch": 0.04548933843630399,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 4.907087460240999e-06,
|
|
"loss": 0.4150247275829315,
|
|
"mean_token_accuracy": 0.8599706739187241,
|
|
"num_tokens": 850751.0,
|
|
"step": 104
|
|
},
|
|
{
|
|
"entropy": 0.5654869377613068,
|
|
"epoch": 0.045926735921268454,
|
|
"grad_norm": 3.109375,
|
|
"learning_rate": 4.9022933048627496e-06,
|
|
"loss": 0.6229053735733032,
|
|
"mean_token_accuracy": 0.7930107563734055,
|
|
"num_tokens": 858943.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 0.4413799047470093,
|
|
"epoch": 0.04636413340623292,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 4.897381017372462e-06,
|
|
"loss": 0.45984703302383423,
|
|
"mean_token_accuracy": 0.8453079164028168,
|
|
"num_tokens": 867135.0,
|
|
"step": 106
|
|
},
|
|
{
|
|
"entropy": 0.4523310959339142,
|
|
"epoch": 0.046801530891197374,
|
|
"grad_norm": 2.53125,
|
|
"learning_rate": 4.8923508393305224e-06,
|
|
"loss": 0.4898549020290375,
|
|
"mean_token_accuracy": 0.8416422307491302,
|
|
"num_tokens": 875327.0,
|
|
"step": 107
|
|
},
|
|
{
|
|
"entropy": 0.4744609370827675,
|
|
"epoch": 0.04723892837616184,
|
|
"grad_norm": 2.765625,
|
|
"learning_rate": 4.887203018094552e-06,
|
|
"loss": 0.5368300676345825,
|
|
"mean_token_accuracy": 0.8267350792884827,
|
|
"num_tokens": 883519.0,
|
|
"step": 108
|
|
},
|
|
{
|
|
"entropy": 0.5006208494305611,
|
|
"epoch": 0.0476763258611263,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 4.881937806807241e-06,
|
|
"loss": 0.5572711229324341,
|
|
"mean_token_accuracy": 0.821480929851532,
|
|
"num_tokens": 891711.0,
|
|
"step": 109
|
|
},
|
|
{
|
|
"entropy": 0.43279464542865753,
|
|
"epoch": 0.04811372334609076,
|
|
"grad_norm": 2.859375,
|
|
"learning_rate": 4.876555464383908e-06,
|
|
"loss": 0.4780406355857849,
|
|
"mean_token_accuracy": 0.8428641259670258,
|
|
"num_tokens": 899903.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 0.49086206406354904,
|
|
"epoch": 0.04855112083105522,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 4.871056255499758e-06,
|
|
"loss": 0.5436443090438843,
|
|
"mean_token_accuracy": 0.8207478076219559,
|
|
"num_tokens": 908095.0,
|
|
"step": 111
|
|
},
|
|
{
|
|
"entropy": 0.42109110951423645,
|
|
"epoch": 0.048988518316019684,
|
|
"grad_norm": 2.6875,
|
|
"learning_rate": 4.8654404505768735e-06,
|
|
"loss": 0.48846113681793213,
|
|
"mean_token_accuracy": 0.8423753827810287,
|
|
"num_tokens": 916287.0,
|
|
"step": 112
|
|
},
|
|
{
|
|
"entropy": 0.45168935507535934,
|
|
"epoch": 0.04942591580098414,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 4.859708325770919e-06,
|
|
"loss": 0.5054396390914917,
|
|
"mean_token_accuracy": 0.8401759564876556,
|
|
"num_tokens": 924479.0,
|
|
"step": 113
|
|
},
|
|
{
|
|
"entropy": 0.4531315267086029,
|
|
"epoch": 0.049863313285948604,
|
|
"grad_norm": 2.71875,
|
|
"learning_rate": 4.8538601629575525e-06,
|
|
"loss": 0.504815936088562,
|
|
"mean_token_accuracy": 0.8376099616289139,
|
|
"num_tokens": 932671.0,
|
|
"step": 114
|
|
},
|
|
{
|
|
"entropy": 0.48239656537771225,
|
|
"epoch": 0.05030071077091307,
|
|
"grad_norm": 2.625,
|
|
"learning_rate": 4.84789624971857e-06,
|
|
"loss": 0.5212786197662354,
|
|
"mean_token_accuracy": 0.8241691142320633,
|
|
"num_tokens": 940863.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 0.4362593963742256,
|
|
"epoch": 0.05073810825587753,
|
|
"grad_norm": 2.5625,
|
|
"learning_rate": 4.841816879327764e-06,
|
|
"loss": 0.4883577823638916,
|
|
"mean_token_accuracy": 0.8366324603557587,
|
|
"num_tokens": 949055.0,
|
|
"step": 116
|
|
},
|
|
{
|
|
"entropy": 0.4954916387796402,
|
|
"epoch": 0.05117550574084199,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 4.835622350736499e-06,
|
|
"loss": 0.5549702048301697,
|
|
"mean_token_accuracy": 0.8211143612861633,
|
|
"num_tokens": 957247.0,
|
|
"step": 117
|
|
},
|
|
{
|
|
"entropy": 0.457555316388607,
|
|
"epoch": 0.05161290322580645,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 4.8293129685590164e-06,
|
|
"loss": 0.5015179514884949,
|
|
"mean_token_accuracy": 0.832722395658493,
|
|
"num_tokens": 965439.0,
|
|
"step": 118
|
|
},
|
|
{
|
|
"entropy": 0.4725177064538002,
|
|
"epoch": 0.052050300710770915,
|
|
"grad_norm": 2.546875,
|
|
"learning_rate": 4.822889043057446e-06,
|
|
"loss": 0.5209950804710388,
|
|
"mean_token_accuracy": 0.8273460417985916,
|
|
"num_tokens": 973631.0,
|
|
"step": 119
|
|
},
|
|
{
|
|
"entropy": 0.4489220380783081,
|
|
"epoch": 0.05248769819573537,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 4.816350890126556e-06,
|
|
"loss": 0.4854542315006256,
|
|
"mean_token_accuracy": 0.8404203355312347,
|
|
"num_tokens": 981823.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 0.47007766366004944,
|
|
"epoch": 0.052925095680699835,
|
|
"grad_norm": 2.671875,
|
|
"learning_rate": 4.809698831278217e-06,
|
|
"loss": 0.5187719464302063,
|
|
"mean_token_accuracy": 0.8310117423534393,
|
|
"num_tokens": 990015.0,
|
|
"step": 121
|
|
},
|
|
{
|
|
"entropy": 0.4113807678222656,
|
|
"epoch": 0.0533624931656643,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 4.802933193625593e-06,
|
|
"loss": 0.4729251563549042,
|
|
"mean_token_accuracy": 0.8432306945323944,
|
|
"num_tokens": 998207.0,
|
|
"step": 122
|
|
},
|
|
{
|
|
"entropy": 0.40296216309070587,
|
|
"epoch": 0.05379989065062876,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 4.796054309867053e-06,
|
|
"loss": 0.4630545973777771,
|
|
"mean_token_accuracy": 0.8541055619716644,
|
|
"num_tokens": 1006399.0,
|
|
"step": 123
|
|
},
|
|
{
|
|
"entropy": 0.5370416194200516,
|
|
"epoch": 0.05423728813559322,
|
|
"grad_norm": 2.890625,
|
|
"learning_rate": 4.789062518269812e-06,
|
|
"loss": 0.6142182350158691,
|
|
"mean_token_accuracy": 0.8086510300636292,
|
|
"num_tokens": 1014591.0,
|
|
"step": 124
|
|
},
|
|
{
|
|
"entropy": 0.41779524832963943,
|
|
"epoch": 0.05467468562055768,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 4.781958162653298e-06,
|
|
"loss": 0.4485539197921753,
|
|
"mean_token_accuracy": 0.851783961057663,
|
|
"num_tokens": 1022783.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 0.47237853705883026,
|
|
"epoch": 0.055112083105522146,
|
|
"grad_norm": 2.5625,
|
|
"learning_rate": 4.774741592372242e-06,
|
|
"loss": 0.5322825908660889,
|
|
"mean_token_accuracy": 0.8250244408845901,
|
|
"num_tokens": 1030975.0,
|
|
"step": 126
|
|
},
|
|
{
|
|
"entropy": 0.5153854340314865,
|
|
"epoch": 0.0555494805904866,
|
|
"grad_norm": 2.65625,
|
|
"learning_rate": 4.767413162299501e-06,
|
|
"loss": 0.5750604271888733,
|
|
"mean_token_accuracy": 0.8148826956748962,
|
|
"num_tokens": 1039167.0,
|
|
"step": 127
|
|
},
|
|
{
|
|
"entropy": 0.48488766700029373,
|
|
"epoch": 0.055986878075451066,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 4.759973232808609e-06,
|
|
"loss": 0.5179052352905273,
|
|
"mean_token_accuracy": 0.8273460417985916,
|
|
"num_tokens": 1047359.0,
|
|
"step": 128
|
|
},
|
|
{
|
|
"entropy": 0.45713143050670624,
|
|
"epoch": 0.05642427556041553,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 4.752422169756048e-06,
|
|
"loss": 0.4798344373703003,
|
|
"mean_token_accuracy": 0.8387096673250198,
|
|
"num_tokens": 1055551.0,
|
|
"step": 129
|
|
},
|
|
{
|
|
"entropy": 0.5863286033272743,
|
|
"epoch": 0.056861673045379986,
|
|
"grad_norm": 2.96875,
|
|
"learning_rate": 4.744760344463267e-06,
|
|
"loss": 0.6585489511489868,
|
|
"mean_token_accuracy": 0.7889784872531891,
|
|
"num_tokens": 1063743.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 0.5283961147069931,
|
|
"epoch": 0.05729907053034445,
|
|
"grad_norm": 2.546875,
|
|
"learning_rate": 4.736988133698416e-06,
|
|
"loss": 0.5767723917961121,
|
|
"mean_token_accuracy": 0.8143939226865768,
|
|
"num_tokens": 1071935.0,
|
|
"step": 131
|
|
},
|
|
{
|
|
"entropy": 0.4788113385438919,
|
|
"epoch": 0.05773646801530891,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 4.729105919657818e-06,
|
|
"loss": 0.5062345862388611,
|
|
"mean_token_accuracy": 0.8274682313203812,
|
|
"num_tokens": 1080127.0,
|
|
"step": 132
|
|
},
|
|
{
|
|
"entropy": 0.4851013496518135,
|
|
"epoch": 0.058173865500273376,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 4.721114089947181e-06,
|
|
"loss": 0.5487649440765381,
|
|
"mean_token_accuracy": 0.8216031342744827,
|
|
"num_tokens": 1088319.0,
|
|
"step": 133
|
|
},
|
|
{
|
|
"entropy": 0.3871802017092705,
|
|
"epoch": 0.05861126298523783,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 4.713013037562531e-06,
|
|
"loss": 0.4369054436683655,
|
|
"mean_token_accuracy": 0.8519765138626099,
|
|
"num_tokens": 1096268.0,
|
|
"step": 134
|
|
},
|
|
{
|
|
"entropy": 0.47404175996780396,
|
|
"epoch": 0.0590486604702023,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 4.704803160870888e-06,
|
|
"loss": 0.5088684558868408,
|
|
"mean_token_accuracy": 0.8313782960176468,
|
|
"num_tokens": 1104460.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 0.578920029103756,
|
|
"epoch": 0.05948605795516676,
|
|
"grad_norm": 2.71875,
|
|
"learning_rate": 4.69648486359068e-06,
|
|
"loss": 0.635509729385376,
|
|
"mean_token_accuracy": 0.8016862124204636,
|
|
"num_tokens": 1112652.0,
|
|
"step": 136
|
|
},
|
|
{
|
|
"entropy": 0.5153625160455704,
|
|
"epoch": 0.05992345544013122,
|
|
"grad_norm": 2.5625,
|
|
"learning_rate": 4.6880585547718845e-06,
|
|
"loss": 0.5665484070777893,
|
|
"mean_token_accuracy": 0.816837728023529,
|
|
"num_tokens": 1120844.0,
|
|
"step": 137
|
|
},
|
|
{
|
|
"entropy": 0.49980155378580093,
|
|
"epoch": 0.06036085292509568,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 4.67952464877592e-06,
|
|
"loss": 0.5389302968978882,
|
|
"mean_token_accuracy": 0.8242912888526917,
|
|
"num_tokens": 1129036.0,
|
|
"step": 138
|
|
},
|
|
{
|
|
"entropy": 0.550069123506546,
|
|
"epoch": 0.060798250410060144,
|
|
"grad_norm": 2.890625,
|
|
"learning_rate": 4.670883565255264e-06,
|
|
"loss": 0.6364497542381287,
|
|
"mean_token_accuracy": 0.8036412447690964,
|
|
"num_tokens": 1137228.0,
|
|
"step": 139
|
|
},
|
|
{
|
|
"entropy": 0.4852010905742645,
|
|
"epoch": 0.0612356478950246,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 4.662135729132821e-06,
|
|
"loss": 0.5219293832778931,
|
|
"mean_token_accuracy": 0.8240469396114349,
|
|
"num_tokens": 1145420.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 0.4792797490954399,
|
|
"epoch": 0.061673045379989064,
|
|
"grad_norm": 2.5625,
|
|
"learning_rate": 4.653281570581023e-06,
|
|
"loss": 0.5293768048286438,
|
|
"mean_token_accuracy": 0.8225806355476379,
|
|
"num_tokens": 1153612.0,
|
|
"step": 141
|
|
},
|
|
{
|
|
"entropy": 0.48140496760606766,
|
|
"epoch": 0.06211044286495353,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 4.644321525000681e-06,
|
|
"loss": 0.5096234679222107,
|
|
"mean_token_accuracy": 0.834921807050705,
|
|
"num_tokens": 1161804.0,
|
|
"step": 142
|
|
},
|
|
{
|
|
"entropy": 0.49927058815956116,
|
|
"epoch": 0.06254784034991799,
|
|
"grad_norm": 2.625,
|
|
"learning_rate": 4.635256032999569e-06,
|
|
"loss": 0.5446697473526001,
|
|
"mean_token_accuracy": 0.8187927603721619,
|
|
"num_tokens": 1169996.0,
|
|
"step": 143
|
|
},
|
|
{
|
|
"entropy": 0.4959957152605057,
|
|
"epoch": 0.06298523783488245,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 4.6260855403707625e-06,
|
|
"loss": 0.5632919669151306,
|
|
"mean_token_accuracy": 0.8185483813285828,
|
|
"num_tokens": 1178188.0,
|
|
"step": 144
|
|
},
|
|
{
|
|
"entropy": 0.49702586978673935,
|
|
"epoch": 0.0634226353198469,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 4.6168104980707105e-06,
|
|
"loss": 0.5302911996841431,
|
|
"mean_token_accuracy": 0.826979473233223,
|
|
"num_tokens": 1186380.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 0.5061358213424683,
|
|
"epoch": 0.06386003280481137,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 4.607431362197067e-06,
|
|
"loss": 0.5666874647140503,
|
|
"mean_token_accuracy": 0.8173264861106873,
|
|
"num_tokens": 1194572.0,
|
|
"step": 146
|
|
},
|
|
{
|
|
"entropy": 0.43493766337633133,
|
|
"epoch": 0.06429743028977583,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 4.597948593966256e-06,
|
|
"loss": 0.44672006368637085,
|
|
"mean_token_accuracy": 0.8514173924922943,
|
|
"num_tokens": 1202764.0,
|
|
"step": 147
|
|
},
|
|
{
|
|
"entropy": 0.48091938346624374,
|
|
"epoch": 0.0647348277747403,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 4.5883626596907945e-06,
|
|
"loss": 0.5102735757827759,
|
|
"mean_token_accuracy": 0.836754634976387,
|
|
"num_tokens": 1210956.0,
|
|
"step": 148
|
|
},
|
|
{
|
|
"entropy": 0.44341713935136795,
|
|
"epoch": 0.06517222525970476,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 4.578674030756364e-06,
|
|
"loss": 0.46256667375564575,
|
|
"mean_token_accuracy": 0.8445747792720795,
|
|
"num_tokens": 1219148.0,
|
|
"step": 149
|
|
},
|
|
{
|
|
"entropy": 0.45063044130802155,
|
|
"epoch": 0.06560962274466922,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 4.568883183598623e-06,
|
|
"loss": 0.4687718152999878,
|
|
"mean_token_accuracy": 0.8429749459028244,
|
|
"num_tokens": 1227312.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 0.46204202622175217,
|
|
"epoch": 0.06604702022963369,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 4.558990599679787e-06,
|
|
"loss": 0.5018566846847534,
|
|
"mean_token_accuracy": 0.831867054104805,
|
|
"num_tokens": 1235504.0,
|
|
"step": 151
|
|
},
|
|
{
|
|
"entropy": 0.44746626913547516,
|
|
"epoch": 0.06648441771459813,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 4.548996765464947e-06,
|
|
"loss": 0.4644359350204468,
|
|
"mean_token_accuracy": 0.8445747792720795,
|
|
"num_tokens": 1243696.0,
|
|
"step": 152
|
|
},
|
|
{
|
|
"entropy": 0.4651805981993675,
|
|
"epoch": 0.0669218151995626,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 4.538902172398151e-06,
|
|
"loss": 0.5121830105781555,
|
|
"mean_token_accuracy": 0.8330889493227005,
|
|
"num_tokens": 1251888.0,
|
|
"step": 153
|
|
},
|
|
{
|
|
"entropy": 0.4517085552215576,
|
|
"epoch": 0.06735921268452706,
|
|
"grad_norm": 2.5625,
|
|
"learning_rate": 4.528707316878233e-06,
|
|
"loss": 0.511009693145752,
|
|
"mean_token_accuracy": 0.837854340672493,
|
|
"num_tokens": 1260080.0,
|
|
"step": 154
|
|
},
|
|
{
|
|
"entropy": 0.4772433340549469,
|
|
"epoch": 0.06779661016949153,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 4.518412700234407e-06,
|
|
"loss": 0.523794412612915,
|
|
"mean_token_accuracy": 0.8319481462240219,
|
|
"num_tokens": 1268224.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 0.4500661939382553,
|
|
"epoch": 0.06823400765445599,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 4.508018828701613e-06,
|
|
"loss": 0.4889807105064392,
|
|
"mean_token_accuracy": 0.8389540612697601,
|
|
"num_tokens": 1276416.0,
|
|
"step": 156
|
|
},
|
|
{
|
|
"entropy": 0.5382272824645042,
|
|
"epoch": 0.06867140513942045,
|
|
"grad_norm": 2.671875,
|
|
"learning_rate": 4.4975262133956235e-06,
|
|
"loss": 0.6048404574394226,
|
|
"mean_token_accuracy": 0.8051075339317322,
|
|
"num_tokens": 1284608.0,
|
|
"step": 157
|
|
},
|
|
{
|
|
"entropy": 0.4137018471956253,
|
|
"epoch": 0.06910880262438492,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 4.486935370287907e-06,
|
|
"loss": 0.4585399627685547,
|
|
"mean_token_accuracy": 0.8515395969152451,
|
|
"num_tokens": 1292800.0,
|
|
"step": 158
|
|
},
|
|
{
|
|
"entropy": 0.5151834562420845,
|
|
"epoch": 0.06954620010934937,
|
|
"grad_norm": 2.78125,
|
|
"learning_rate": 4.476246820180259e-06,
|
|
"loss": 0.5867331027984619,
|
|
"mean_token_accuracy": 0.8187927752733231,
|
|
"num_tokens": 1300992.0,
|
|
"step": 159
|
|
},
|
|
{
|
|
"entropy": 0.4778127670288086,
|
|
"epoch": 0.06998359759431383,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 4.465461088679189e-06,
|
|
"loss": 0.5396682024002075,
|
|
"mean_token_accuracy": 0.8280728608369827,
|
|
"num_tokens": 1309112.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 0.5015709027647972,
|
|
"epoch": 0.07042099507927829,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 4.454578706170075e-06,
|
|
"loss": 0.5446302890777588,
|
|
"mean_token_accuracy": 0.8217253088951111,
|
|
"num_tokens": 1317304.0,
|
|
"step": 161
|
|
},
|
|
{
|
|
"entropy": 0.4737255424261093,
|
|
"epoch": 0.07085839256424276,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 4.44360020779108e-06,
|
|
"loss": 0.5138781666755676,
|
|
"mean_token_accuracy": 0.829912006855011,
|
|
"num_tokens": 1325496.0,
|
|
"step": 162
|
|
},
|
|
{
|
|
"entropy": 0.49487005919218063,
|
|
"epoch": 0.07129579004920722,
|
|
"grad_norm": 2.53125,
|
|
"learning_rate": 4.432526133406843e-06,
|
|
"loss": 0.5455970764160156,
|
|
"mean_token_accuracy": 0.8183040022850037,
|
|
"num_tokens": 1333688.0,
|
|
"step": 163
|
|
},
|
|
{
|
|
"entropy": 0.4824206978082657,
|
|
"epoch": 0.07173318753417168,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 4.421357027581921e-06,
|
|
"loss": 0.5247961282730103,
|
|
"mean_token_accuracy": 0.8318670690059662,
|
|
"num_tokens": 1341880.0,
|
|
"step": 164
|
|
},
|
|
{
|
|
"entropy": 0.5007349699735641,
|
|
"epoch": 0.07217058501913615,
|
|
"grad_norm": 2.625,
|
|
"learning_rate": 4.410093439554019e-06,
|
|
"loss": 0.5320008993148804,
|
|
"mean_token_accuracy": 0.8262404948472977,
|
|
"num_tokens": 1349828.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 0.45331911742687225,
|
|
"epoch": 0.0726079825041006,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 4.398735923206978e-06,
|
|
"loss": 0.5013130903244019,
|
|
"mean_token_accuracy": 0.8371212184429169,
|
|
"num_tokens": 1358020.0,
|
|
"step": 166
|
|
},
|
|
{
|
|
"entropy": 0.5055417120456696,
|
|
"epoch": 0.07304537998906506,
|
|
"grad_norm": 2.5625,
|
|
"learning_rate": 4.38728503704354e-06,
|
|
"loss": 0.5585504174232483,
|
|
"mean_token_accuracy": 0.820992186665535,
|
|
"num_tokens": 1366212.0,
|
|
"step": 167
|
|
},
|
|
{
|
|
"entropy": 0.5267198234796524,
|
|
"epoch": 0.07348277747402952,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 4.375741344157879e-06,
|
|
"loss": 0.5661095976829529,
|
|
"mean_token_accuracy": 0.8141495585441589,
|
|
"num_tokens": 1374404.0,
|
|
"step": 168
|
|
},
|
|
{
|
|
"entropy": 0.4898260533809662,
|
|
"epoch": 0.07392017495899399,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 4.364105412207914e-06,
|
|
"loss": 0.5348383188247681,
|
|
"mean_token_accuracy": 0.8258797526359558,
|
|
"num_tokens": 1382596.0,
|
|
"step": 169
|
|
},
|
|
{
|
|
"entropy": 0.5083483383059502,
|
|
"epoch": 0.07435757244395845,
|
|
"grad_norm": 2.515625,
|
|
"learning_rate": 4.352377813387398e-06,
|
|
"loss": 0.5432155132293701,
|
|
"mean_token_accuracy": 0.8250244408845901,
|
|
"num_tokens": 1390788.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 0.44939737766981125,
|
|
"epoch": 0.07479496992892291,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 4.340559124397774e-06,
|
|
"loss": 0.500606119632721,
|
|
"mean_token_accuracy": 0.8351661711931229,
|
|
"num_tokens": 1398980.0,
|
|
"step": 171
|
|
},
|
|
{
|
|
"entropy": 0.44741733372211456,
|
|
"epoch": 0.07523236741388736,
|
|
"grad_norm": 2.546875,
|
|
"learning_rate": 4.32864992641982e-06,
|
|
"loss": 0.4959651231765747,
|
|
"mean_token_accuracy": 0.8425872474908829,
|
|
"num_tokens": 1407145.0,
|
|
"step": 172
|
|
},
|
|
{
|
|
"entropy": 0.5106001943349838,
|
|
"epoch": 0.07566976489885183,
|
|
"grad_norm": 2.59375,
|
|
"learning_rate": 4.316650805085068e-06,
|
|
"loss": 0.5437954068183899,
|
|
"mean_token_accuracy": 0.821130782365799,
|
|
"num_tokens": 1415243.0,
|
|
"step": 173
|
|
},
|
|
{
|
|
"entropy": 0.5708731412887573,
|
|
"epoch": 0.07610716238381629,
|
|
"grad_norm": 2.671875,
|
|
"learning_rate": 4.304562350447008e-06,
|
|
"loss": 0.6295499801635742,
|
|
"mean_token_accuracy": 0.8030303120613098,
|
|
"num_tokens": 1423435.0,
|
|
"step": 174
|
|
},
|
|
{
|
|
"entropy": 0.45462463796138763,
|
|
"epoch": 0.07654455986878075,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 4.292385156952069e-06,
|
|
"loss": 0.48111745715141296,
|
|
"mean_token_accuracy": 0.8400537669658661,
|
|
"num_tokens": 1431627.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 0.6211910545825958,
|
|
"epoch": 0.07698195735374522,
|
|
"grad_norm": 2.625,
|
|
"learning_rate": 4.2801198234103895e-06,
|
|
"loss": 0.669766902923584,
|
|
"mean_token_accuracy": 0.7941104620695114,
|
|
"num_tokens": 1439819.0,
|
|
"step": 176
|
|
},
|
|
{
|
|
"entropy": 0.4480922445654869,
|
|
"epoch": 0.07741935483870968,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 4.267766952966369e-06,
|
|
"loss": 0.4825879633426666,
|
|
"mean_token_accuracy": 0.8418866097927094,
|
|
"num_tokens": 1448011.0,
|
|
"step": 177
|
|
},
|
|
{
|
|
"entropy": 0.5084129646420479,
|
|
"epoch": 0.07785675232367414,
|
|
"grad_norm": 2.734375,
|
|
"learning_rate": 4.255327153069014e-06,
|
|
"loss": 0.574237048625946,
|
|
"mean_token_accuracy": 0.8087732046842575,
|
|
"num_tokens": 1456203.0,
|
|
"step": 178
|
|
},
|
|
{
|
|
"entropy": 0.4354983791708946,
|
|
"epoch": 0.0782941498086386,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 4.242801035442059e-06,
|
|
"loss": 0.47676539421081543,
|
|
"mean_token_accuracy": 0.8424975723028183,
|
|
"num_tokens": 1464395.0,
|
|
"step": 179
|
|
},
|
|
{
|
|
"entropy": 0.5039181187748909,
|
|
"epoch": 0.07873154729360306,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 4.2301892160538895e-06,
|
|
"loss": 0.5764544010162354,
|
|
"mean_token_accuracy": 0.8211143761873245,
|
|
"num_tokens": 1472587.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 0.4349247217178345,
|
|
"epoch": 0.07916894477856752,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 4.217492315087255e-06,
|
|
"loss": 0.4684079885482788,
|
|
"mean_token_accuracy": 0.8488514274358749,
|
|
"num_tokens": 1480779.0,
|
|
"step": 181
|
|
},
|
|
{
|
|
"entropy": 0.529822438955307,
|
|
"epoch": 0.07960634226353198,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 4.204710956908763e-06,
|
|
"loss": 0.592373251914978,
|
|
"mean_token_accuracy": 0.8134164214134216,
|
|
"num_tokens": 1488971.0,
|
|
"step": 182
|
|
},
|
|
{
|
|
"entropy": 0.5234254226088524,
|
|
"epoch": 0.08004373974849645,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 4.191845770038186e-06,
|
|
"loss": 0.5723234415054321,
|
|
"mean_token_accuracy": 0.8096285313367844,
|
|
"num_tokens": 1497163.0,
|
|
"step": 183
|
|
},
|
|
{
|
|
"entropy": 0.5915178805589676,
|
|
"epoch": 0.08048113723346091,
|
|
"grad_norm": 2.6875,
|
|
"learning_rate": 4.178897387117547e-06,
|
|
"loss": 0.6336113214492798,
|
|
"mean_token_accuracy": 0.7989980429410934,
|
|
"num_tokens": 1505355.0,
|
|
"step": 184
|
|
},
|
|
{
|
|
"entropy": 0.5129002556204796,
|
|
"epoch": 0.08091853471842538,
|
|
"grad_norm": 2.53125,
|
|
"learning_rate": 4.1658664448800105e-06,
|
|
"loss": 0.5581027269363403,
|
|
"mean_token_accuracy": 0.8189149498939514,
|
|
"num_tokens": 1513547.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 0.5568132996559143,
|
|
"epoch": 0.08135593220338982,
|
|
"grad_norm": 2.546875,
|
|
"learning_rate": 4.152753584118576e-06,
|
|
"loss": 0.5956103801727295,
|
|
"mean_token_accuracy": 0.8057184815406799,
|
|
"num_tokens": 1521739.0,
|
|
"step": 186
|
|
},
|
|
{
|
|
"entropy": 0.5218602269887924,
|
|
"epoch": 0.08179332968835429,
|
|
"grad_norm": 2.546875,
|
|
"learning_rate": 4.139559449654561e-06,
|
|
"loss": 0.5501708984375,
|
|
"mean_token_accuracy": 0.8205034136772156,
|
|
"num_tokens": 1529931.0,
|
|
"step": 187
|
|
},
|
|
{
|
|
"entropy": 0.48786159604787827,
|
|
"epoch": 0.08223072717331875,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 4.1262846903058966e-06,
|
|
"loss": 0.5217365622520447,
|
|
"mean_token_accuracy": 0.8286901414394379,
|
|
"num_tokens": 1538123.0,
|
|
"step": 188
|
|
},
|
|
{
|
|
"entropy": 0.5472189486026764,
|
|
"epoch": 0.08266812465828322,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 4.11292995885522e-06,
|
|
"loss": 0.5948308706283569,
|
|
"mean_token_accuracy": 0.8060850501060486,
|
|
"num_tokens": 1546315.0,
|
|
"step": 189
|
|
},
|
|
{
|
|
"entropy": 0.4940583184361458,
|
|
"epoch": 0.08310552214324768,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 4.099495912017773e-06,
|
|
"loss": 0.5291944742202759,
|
|
"mean_token_accuracy": 0.8277125954627991,
|
|
"num_tokens": 1554507.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 0.5450269728899002,
|
|
"epoch": 0.08354291962821214,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 4.085983210409114e-06,
|
|
"loss": 0.5771312117576599,
|
|
"mean_token_accuracy": 0.8102394938468933,
|
|
"num_tokens": 1562699.0,
|
|
"step": 191
|
|
},
|
|
{
|
|
"entropy": 0.4809708967804909,
|
|
"epoch": 0.0839803171131766,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 4.072392518512623e-06,
|
|
"loss": 0.515668511390686,
|
|
"mean_token_accuracy": 0.8312561064958572,
|
|
"num_tokens": 1570891.0,
|
|
"step": 192
|
|
},
|
|
{
|
|
"entropy": 0.5454308390617371,
|
|
"epoch": 0.08441771459814106,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 4.058724504646834e-06,
|
|
"loss": 0.5704600811004639,
|
|
"mean_token_accuracy": 0.8162267804145813,
|
|
"num_tokens": 1579083.0,
|
|
"step": 193
|
|
},
|
|
{
|
|
"entropy": 0.46287182718515396,
|
|
"epoch": 0.08485511208310552,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 4.044979840932567e-06,
|
|
"loss": 0.4835647940635681,
|
|
"mean_token_accuracy": 0.8400537520647049,
|
|
"num_tokens": 1587275.0,
|
|
"step": 194
|
|
},
|
|
{
|
|
"entropy": 0.4392080307006836,
|
|
"epoch": 0.08529250956806998,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 4.031159203259876e-06,
|
|
"loss": 0.47868117690086365,
|
|
"mean_token_accuracy": 0.8413367718458176,
|
|
"num_tokens": 1595298.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 0.4842827618122101,
|
|
"epoch": 0.08572990705303445,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 4.0172632712548145e-06,
|
|
"loss": 0.5347608327865601,
|
|
"mean_token_accuracy": 0.8262463212013245,
|
|
"num_tokens": 1603490.0,
|
|
"step": 196
|
|
},
|
|
{
|
|
"entropy": 0.42982233315706253,
|
|
"epoch": 0.08616730453799891,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 4.003292728246015e-06,
|
|
"loss": 0.4687855839729309,
|
|
"mean_token_accuracy": 0.8500733077526093,
|
|
"num_tokens": 1611682.0,
|
|
"step": 197
|
|
},
|
|
{
|
|
"entropy": 0.47412050515413284,
|
|
"epoch": 0.08660470202296337,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 3.989248261231084e-06,
|
|
"loss": 0.4940436780452728,
|
|
"mean_token_accuracy": 0.8384652882814407,
|
|
"num_tokens": 1619874.0,
|
|
"step": 198
|
|
},
|
|
{
|
|
"entropy": 0.4639420583844185,
|
|
"epoch": 0.08704209950792782,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 3.975130560842821e-06,
|
|
"loss": 0.4897710680961609,
|
|
"mean_token_accuracy": 0.8346655517816544,
|
|
"num_tokens": 1628017.0,
|
|
"step": 199
|
|
},
|
|
{
|
|
"entropy": 0.4265002906322479,
|
|
"epoch": 0.08747949699289229,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 3.960940321315257e-06,
|
|
"loss": 0.4600346088409424,
|
|
"mean_token_accuracy": 0.8492179811000824,
|
|
"num_tokens": 1636209.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 0.3981725126504898,
|
|
"epoch": 0.08791689447785675,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 3.946678240449515e-06,
|
|
"loss": 0.43623173236846924,
|
|
"mean_token_accuracy": 0.8528836667537689,
|
|
"num_tokens": 1644401.0,
|
|
"step": 201
|
|
},
|
|
{
|
|
"entropy": 0.5264689773321152,
|
|
"epoch": 0.08835429196282121,
|
|
"grad_norm": 2.828125,
|
|
"learning_rate": 3.9323450195794954e-06,
|
|
"loss": 0.5827093720436096,
|
|
"mean_token_accuracy": 0.8156800270080566,
|
|
"num_tokens": 1652489.0,
|
|
"step": 202
|
|
},
|
|
{
|
|
"entropy": 0.4846732020378113,
|
|
"epoch": 0.08879168944778568,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 3.9179413635373895e-06,
|
|
"loss": 0.5261396169662476,
|
|
"mean_token_accuracy": 0.8269794881343842,
|
|
"num_tokens": 1660681.0,
|
|
"step": 203
|
|
},
|
|
{
|
|
"entropy": 0.5061644613742828,
|
|
"epoch": 0.08922908693275014,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 3.9034679806190204e-06,
|
|
"loss": 0.5423262119293213,
|
|
"mean_token_accuracy": 0.8249022364616394,
|
|
"num_tokens": 1668873.0,
|
|
"step": 204
|
|
},
|
|
{
|
|
"entropy": 0.5438003838062286,
|
|
"epoch": 0.0896664844177146,
|
|
"grad_norm": 2.5625,
|
|
"learning_rate": 3.888925582549006e-06,
|
|
"loss": 0.5897356867790222,
|
|
"mean_token_accuracy": 0.8076735138893127,
|
|
"num_tokens": 1677065.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 0.4519123062491417,
|
|
"epoch": 0.09010388190267905,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 3.874314884445771e-06,
|
|
"loss": 0.4760674238204956,
|
|
"mean_token_accuracy": 0.8390762358903885,
|
|
"num_tokens": 1685257.0,
|
|
"step": 206
|
|
},
|
|
{
|
|
"entropy": 0.5100989043712616,
|
|
"epoch": 0.09054127938764352,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 3.859636604786372e-06,
|
|
"loss": 0.5384048223495483,
|
|
"mean_token_accuracy": 0.8235581666231155,
|
|
"num_tokens": 1693449.0,
|
|
"step": 207
|
|
},
|
|
{
|
|
"entropy": 0.4979291707277298,
|
|
"epoch": 0.09097867687260798,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 3.8448914653711725e-06,
|
|
"loss": 0.5454040169715881,
|
|
"mean_token_accuracy": 0.82233627140522,
|
|
"num_tokens": 1701641.0,
|
|
"step": 208
|
|
},
|
|
{
|
|
"entropy": 0.48866020888090134,
|
|
"epoch": 0.09141607435757244,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 3.830080191288342e-06,
|
|
"loss": 0.5353090167045593,
|
|
"mean_token_accuracy": 0.8250244408845901,
|
|
"num_tokens": 1709833.0,
|
|
"step": 209
|
|
},
|
|
{
|
|
"entropy": 0.4369449093937874,
|
|
"epoch": 0.09185347184253691,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 3.815203510878209e-06,
|
|
"loss": 0.47227001190185547,
|
|
"mean_token_accuracy": 0.8410312831401825,
|
|
"num_tokens": 1718025.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 0.5231678262352943,
|
|
"epoch": 0.09229086932750137,
|
|
"grad_norm": 2.546875,
|
|
"learning_rate": 3.8002621556974367e-06,
|
|
"loss": 0.5623139142990112,
|
|
"mean_token_accuracy": 0.8194037079811096,
|
|
"num_tokens": 1726217.0,
|
|
"step": 211
|
|
},
|
|
{
|
|
"entropy": 0.47718561440706253,
|
|
"epoch": 0.09272826681246583,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 3.785256860483054e-06,
|
|
"loss": 0.5391553640365601,
|
|
"mean_token_accuracy": 0.8256353586912155,
|
|
"num_tokens": 1734409.0,
|
|
"step": 212
|
|
},
|
|
{
|
|
"entropy": 0.48632175475358963,
|
|
"epoch": 0.09316566429743028,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 3.770188363116324e-06,
|
|
"loss": 0.5253764986991882,
|
|
"mean_token_accuracy": 0.8268572837114334,
|
|
"num_tokens": 1742601.0,
|
|
"step": 213
|
|
},
|
|
{
|
|
"entropy": 0.5466743558645248,
|
|
"epoch": 0.09360306178239475,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 3.755057404586455e-06,
|
|
"loss": 0.5637357234954834,
|
|
"mean_token_accuracy": 0.8114613890647888,
|
|
"num_tokens": 1750793.0,
|
|
"step": 214
|
|
},
|
|
{
|
|
"entropy": 0.4229523092508316,
|
|
"epoch": 0.09404045926735921,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 3.7398647289541703e-06,
|
|
"loss": 0.46079084277153015,
|
|
"mean_token_accuracy": 0.8492179811000824,
|
|
"num_tokens": 1758985.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 0.5077691823244095,
|
|
"epoch": 0.09447785675232367,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 3.724611083315109e-06,
|
|
"loss": 0.5458195209503174,
|
|
"mean_token_accuracy": 0.8192094713449478,
|
|
"num_tokens": 1767017.0,
|
|
"step": 216
|
|
},
|
|
{
|
|
"entropy": 0.45738635212183,
|
|
"epoch": 0.09491525423728814,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 3.7092972177630998e-06,
|
|
"loss": 0.49977749586105347,
|
|
"mean_token_accuracy": 0.8341886550188065,
|
|
"num_tokens": 1775209.0,
|
|
"step": 217
|
|
},
|
|
{
|
|
"entropy": 0.5932013764977455,
|
|
"epoch": 0.0953526517222526,
|
|
"grad_norm": 2.515625,
|
|
"learning_rate": 3.6939238853532657e-06,
|
|
"loss": 0.6106526851654053,
|
|
"mean_token_accuracy": 0.8014418482780457,
|
|
"num_tokens": 1783401.0,
|
|
"step": 218
|
|
},
|
|
{
|
|
"entropy": 0.49210917949676514,
|
|
"epoch": 0.09579004920721707,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 3.6784918420649952e-06,
|
|
"loss": 0.5386967658996582,
|
|
"mean_token_accuracy": 0.8255131840705872,
|
|
"num_tokens": 1791593.0,
|
|
"step": 219
|
|
},
|
|
{
|
|
"entropy": 0.4963834807276726,
|
|
"epoch": 0.09622744669218151,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 3.663001846764769e-06,
|
|
"loss": 0.5275002717971802,
|
|
"mean_token_accuracy": 0.8225806504487991,
|
|
"num_tokens": 1799785.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 0.4301696792244911,
|
|
"epoch": 0.09666484417714598,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 3.6474546611688446e-06,
|
|
"loss": 0.46401843428611755,
|
|
"mean_token_accuracy": 0.8456361144781113,
|
|
"num_tokens": 1807918.0,
|
|
"step": 221
|
|
},
|
|
{
|
|
"entropy": 0.456448033452034,
|
|
"epoch": 0.09710224166211044,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 3.631851049805793e-06,
|
|
"loss": 0.49123865365982056,
|
|
"mean_token_accuracy": 0.8367546498775482,
|
|
"num_tokens": 1816110.0,
|
|
"step": 222
|
|
},
|
|
{
|
|
"entropy": 0.5055864825844765,
|
|
"epoch": 0.0975396391470749,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 3.6161917799789076e-06,
|
|
"loss": 0.5433682799339294,
|
|
"mean_token_accuracy": 0.8196480870246887,
|
|
"num_tokens": 1824302.0,
|
|
"step": 223
|
|
},
|
|
{
|
|
"entropy": 0.48989594727754593,
|
|
"epoch": 0.09797703663203937,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 3.600477621728473e-06,
|
|
"loss": 0.5149828195571899,
|
|
"mean_token_accuracy": 0.8290567100048065,
|
|
"num_tokens": 1832494.0,
|
|
"step": 224
|
|
},
|
|
{
|
|
"entropy": 0.461464524269104,
|
|
"epoch": 0.09841443411700383,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 3.5847093477938955e-06,
|
|
"loss": 0.46980535984039307,
|
|
"mean_token_accuracy": 0.843352884054184,
|
|
"num_tokens": 1840686.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 0.5267182663083076,
|
|
"epoch": 0.09885183160196828,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 3.5688877335757055e-06,
|
|
"loss": 0.5791686773300171,
|
|
"mean_token_accuracy": 0.8118279427289963,
|
|
"num_tokens": 1848878.0,
|
|
"step": 226
|
|
},
|
|
{
|
|
"entropy": 0.5815688967704773,
|
|
"epoch": 0.09928922908693275,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 3.553013557097428e-06,
|
|
"loss": 0.6133584380149841,
|
|
"mean_token_accuracy": 0.8036650568246841,
|
|
"num_tokens": 1856831.0,
|
|
"step": 227
|
|
},
|
|
{
|
|
"entropy": 0.4044683054089546,
|
|
"epoch": 0.09972662657189721,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 3.5370875989673198e-06,
|
|
"loss": 0.43988388776779175,
|
|
"mean_token_accuracy": 0.851725623011589,
|
|
"num_tokens": 1864824.0,
|
|
"step": 228
|
|
},
|
|
{
|
|
"entropy": 0.44032787531614304,
|
|
"epoch": 0.10016402405686167,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 3.521110642339991e-06,
|
|
"loss": 0.4767438769340515,
|
|
"mean_token_accuracy": 0.843352884054184,
|
|
"num_tokens": 1873016.0,
|
|
"step": 229
|
|
},
|
|
{
|
|
"entropy": 0.5057295858860016,
|
|
"epoch": 0.10060142154182614,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 3.505083472877884e-06,
|
|
"loss": 0.5249795913696289,
|
|
"mean_token_accuracy": 0.8291788697242737,
|
|
"num_tokens": 1881208.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 0.46067339181900024,
|
|
"epoch": 0.1010388190267906,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 3.4890068787126475e-06,
|
|
"loss": 0.4979941248893738,
|
|
"mean_token_accuracy": 0.8371212184429169,
|
|
"num_tokens": 1889400.0,
|
|
"step": 231
|
|
},
|
|
{
|
|
"entropy": 0.3977827876806259,
|
|
"epoch": 0.10147621651175506,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 3.472881650406373e-06,
|
|
"loss": 0.4426778256893158,
|
|
"mean_token_accuracy": 0.8523949086666107,
|
|
"num_tokens": 1897592.0,
|
|
"step": 232
|
|
},
|
|
{
|
|
"entropy": 0.49042895436286926,
|
|
"epoch": 0.10191361399671951,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 3.4567085809127247e-06,
|
|
"loss": 0.5123292207717896,
|
|
"mean_token_accuracy": 0.8297898322343826,
|
|
"num_tokens": 1905784.0,
|
|
"step": 233
|
|
},
|
|
{
|
|
"entropy": 0.5105108842253685,
|
|
"epoch": 0.10235101148168398,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 3.4404884655379433e-06,
|
|
"loss": 0.5438479781150818,
|
|
"mean_token_accuracy": 0.8196481019258499,
|
|
"num_tokens": 1913976.0,
|
|
"step": 234
|
|
},
|
|
{
|
|
"entropy": 0.4192558228969574,
|
|
"epoch": 0.10278840896664844,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 3.424222101901738e-06,
|
|
"loss": 0.4592979848384857,
|
|
"mean_token_accuracy": 0.8531280606985092,
|
|
"num_tokens": 1922168.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 0.49979063868522644,
|
|
"epoch": 0.1032258064516129,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 3.4079102898980633e-06,
|
|
"loss": 0.5456841588020325,
|
|
"mean_token_accuracy": 0.8219696879386902,
|
|
"num_tokens": 1930360.0,
|
|
"step": 236
|
|
},
|
|
{
|
|
"entropy": 0.42070452123880386,
|
|
"epoch": 0.10366320393657737,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 3.391553831655783e-06,
|
|
"loss": 0.44820719957351685,
|
|
"mean_token_accuracy": 0.8532502502202988,
|
|
"num_tokens": 1938552.0,
|
|
"step": 237
|
|
},
|
|
{
|
|
"entropy": 0.5203416794538498,
|
|
"epoch": 0.10410060142154183,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 3.375153531499231e-06,
|
|
"loss": 0.5365148186683655,
|
|
"mean_token_accuracy": 0.8222140669822693,
|
|
"num_tokens": 1946744.0,
|
|
"step": 238
|
|
},
|
|
{
|
|
"entropy": 0.42554354667663574,
|
|
"epoch": 0.1045379989065063,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 3.358710195908653e-06,
|
|
"loss": 0.44316649436950684,
|
|
"mean_token_accuracy": 0.8528836667537689,
|
|
"num_tokens": 1954936.0,
|
|
"step": 239
|
|
},
|
|
{
|
|
"entropy": 0.49205484241247177,
|
|
"epoch": 0.10497539639147074,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 3.3422246334805504e-06,
|
|
"loss": 0.5142548084259033,
|
|
"mean_token_accuracy": 0.8328445851802826,
|
|
"num_tokens": 1963128.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 0.45800766348838806,
|
|
"epoch": 0.1054127938764352,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 3.3256976548879183e-06,
|
|
"loss": 0.4986850917339325,
|
|
"mean_token_accuracy": 0.8334555178880692,
|
|
"num_tokens": 1971320.0,
|
|
"step": 241
|
|
},
|
|
{
|
|
"entropy": 0.5307494774460793,
|
|
"epoch": 0.10585019136139967,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 3.309130072840379e-06,
|
|
"loss": 0.5546558499336243,
|
|
"mean_token_accuracy": 0.8191862851381302,
|
|
"num_tokens": 1979406.0,
|
|
"step": 242
|
|
},
|
|
{
|
|
"entropy": 0.4573250040411949,
|
|
"epoch": 0.10628758884636413,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 3.292522702044221e-06,
|
|
"loss": 0.4930002689361572,
|
|
"mean_token_accuracy": 0.8427419364452362,
|
|
"num_tokens": 1987598.0,
|
|
"step": 243
|
|
},
|
|
{
|
|
"entropy": 0.5074465125799179,
|
|
"epoch": 0.1067249863313286,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 3.27587635916233e-06,
|
|
"loss": 0.5355194211006165,
|
|
"mean_token_accuracy": 0.8251466304063797,
|
|
"num_tokens": 1995790.0,
|
|
"step": 244
|
|
},
|
|
{
|
|
"entropy": 0.44894687086343765,
|
|
"epoch": 0.10716238381629306,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 3.259191862774037e-06,
|
|
"loss": 0.4787830412387848,
|
|
"mean_token_accuracy": 0.8409090936183929,
|
|
"num_tokens": 2003982.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 0.5497879087924957,
|
|
"epoch": 0.10759978130125752,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 3.242470033334859e-06,
|
|
"loss": 0.605748176574707,
|
|
"mean_token_accuracy": 0.8096285462379456,
|
|
"num_tokens": 2012174.0,
|
|
"step": 246
|
|
},
|
|
{
|
|
"entropy": 0.4496011435985565,
|
|
"epoch": 0.10803717878622197,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 3.225711693136156e-06,
|
|
"loss": 0.47699061036109924,
|
|
"mean_token_accuracy": 0.836754634976387,
|
|
"num_tokens": 2020366.0,
|
|
"step": 247
|
|
},
|
|
{
|
|
"entropy": 0.4718461260199547,
|
|
"epoch": 0.10847457627118644,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 3.2089176662646946e-06,
|
|
"loss": 0.5072636604309082,
|
|
"mean_token_accuracy": 0.8390762507915497,
|
|
"num_tokens": 2028558.0,
|
|
"step": 248
|
|
},
|
|
{
|
|
"entropy": 0.4522283151745796,
|
|
"epoch": 0.1089119737561509,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 3.1920887785621233e-06,
|
|
"loss": 0.48412400484085083,
|
|
"mean_token_accuracy": 0.8443304002285004,
|
|
"num_tokens": 2036750.0,
|
|
"step": 249
|
|
},
|
|
{
|
|
"entropy": 0.48243827372789383,
|
|
"epoch": 0.10934937124111536,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 3.175225857584364e-06,
|
|
"loss": 0.5176639556884766,
|
|
"mean_token_accuracy": 0.8324780166149139,
|
|
"num_tokens": 2044942.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 0.43758024275302887,
|
|
"epoch": 0.10978676872607983,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 3.1583297325609117e-06,
|
|
"loss": 0.4812491536140442,
|
|
"mean_token_accuracy": 0.8454301208257675,
|
|
"num_tokens": 2053134.0,
|
|
"step": 251
|
|
},
|
|
{
|
|
"entropy": 0.5806743204593658,
|
|
"epoch": 0.11022416621104429,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 3.1414012343540667e-06,
|
|
"loss": 0.6051475405693054,
|
|
"mean_token_accuracy": 0.8000977486371994,
|
|
"num_tokens": 2061326.0,
|
|
"step": 252
|
|
},
|
|
{
|
|
"entropy": 0.4813562035560608,
|
|
"epoch": 0.11066156369600874,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 3.1244411954180677e-06,
|
|
"loss": 0.4888189136981964,
|
|
"mean_token_accuracy": 0.8377321511507034,
|
|
"num_tokens": 2069518.0,
|
|
"step": 253
|
|
},
|
|
{
|
|
"entropy": 0.4688641130924225,
|
|
"epoch": 0.1110989611809732,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 3.10745044975816e-06,
|
|
"loss": 0.4884192943572998,
|
|
"mean_token_accuracy": 0.8410312831401825,
|
|
"num_tokens": 2077710.0,
|
|
"step": 254
|
|
},
|
|
{
|
|
"entropy": 0.5002786964178085,
|
|
"epoch": 0.11153635866593767,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 3.0904298328895865e-06,
|
|
"loss": 0.53147292137146,
|
|
"mean_token_accuracy": 0.8268572837114334,
|
|
"num_tokens": 2085902.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 0.5001499354839325,
|
|
"epoch": 0.11197375615090213,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 3.073380181796495e-06,
|
|
"loss": 0.5351055264472961,
|
|
"mean_token_accuracy": 0.8217253237962723,
|
|
"num_tokens": 2094094.0,
|
|
"step": 256
|
|
},
|
|
{
|
|
"epoch": 0.11197375615090213,
|
|
"eval_entropy": 0.4864933586810246,
|
|
"eval_loss": 0.5185804963111877,
|
|
"eval_mean_token_accuracy": 0.8292105375242628,
|
|
"eval_num_tokens": 2094094.0,
|
|
"eval_runtime": 32.8305,
|
|
"eval_samples_per_second": 29.333,
|
|
"eval_steps_per_second": 3.686,
|
|
"step": 256
|
|
},
|
|
{
|
|
"entropy": 0.5339047610759735,
|
|
"epoch": 0.1124111536358666,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 3.056302334890786e-06,
|
|
"loss": 0.5549373030662537,
|
|
"mean_token_accuracy": 0.8195258975028992,
|
|
"num_tokens": 2102286.0,
|
|
"step": 257
|
|
},
|
|
{
|
|
"entropy": 0.5548300221562386,
|
|
"epoch": 0.11284855112083106,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 3.039197131970881e-06,
|
|
"loss": 0.5810383558273315,
|
|
"mean_token_accuracy": 0.8132295906543732,
|
|
"num_tokens": 2110449.0,
|
|
"step": 258
|
|
},
|
|
{
|
|
"entropy": 0.48320460319519043,
|
|
"epoch": 0.11328594860579552,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 3.022065414180425e-06,
|
|
"loss": 0.5107712745666504,
|
|
"mean_token_accuracy": 0.8238025456666946,
|
|
"num_tokens": 2118641.0,
|
|
"step": 259
|
|
},
|
|
{
|
|
"entropy": 0.500133715569973,
|
|
"epoch": 0.11372334609075997,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 3.0049080239669243e-06,
|
|
"loss": 0.5098925232887268,
|
|
"mean_token_accuracy": 0.8319892585277557,
|
|
"num_tokens": 2126833.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 0.4652687683701515,
|
|
"epoch": 0.11416074357572444,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 2.9877258050403214e-06,
|
|
"loss": 0.5014001727104187,
|
|
"mean_token_accuracy": 0.8366324454545975,
|
|
"num_tokens": 2135025.0,
|
|
"step": 261
|
|
},
|
|
{
|
|
"entropy": 0.4709179848432541,
|
|
"epoch": 0.1145981410606889,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 2.9705196023315007e-06,
|
|
"loss": 0.4898775815963745,
|
|
"mean_token_accuracy": 0.8367546498775482,
|
|
"num_tokens": 2143217.0,
|
|
"step": 262
|
|
},
|
|
{
|
|
"entropy": 0.4747740998864174,
|
|
"epoch": 0.11503553854565336,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 2.9532902619507465e-06,
|
|
"loss": 0.5130186676979065,
|
|
"mean_token_accuracy": 0.8319892436265945,
|
|
"num_tokens": 2151409.0,
|
|
"step": 263
|
|
},
|
|
{
|
|
"entropy": 0.4799908697605133,
|
|
"epoch": 0.11547293603061783,
|
|
"grad_norm": 2.546875,
|
|
"learning_rate": 2.9360386311461276e-06,
|
|
"loss": 0.5306564569473267,
|
|
"mean_token_accuracy": 0.826979473233223,
|
|
"num_tokens": 2159601.0,
|
|
"step": 264
|
|
},
|
|
{
|
|
"entropy": 0.4681060016155243,
|
|
"epoch": 0.11591033351558229,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 2.9187655582618413e-06,
|
|
"loss": 0.5099412202835083,
|
|
"mean_token_accuracy": 0.8284457474946976,
|
|
"num_tokens": 2167793.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 0.4734398275613785,
|
|
"epoch": 0.11634773100054675,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 2.901471892696493e-06,
|
|
"loss": 0.49233460426330566,
|
|
"mean_token_accuracy": 0.8350439816713333,
|
|
"num_tokens": 2175985.0,
|
|
"step": 266
|
|
},
|
|
{
|
|
"entropy": 0.5704035013914108,
|
|
"epoch": 0.1167851284855112,
|
|
"grad_norm": 2.65625,
|
|
"learning_rate": 2.8841584848613254e-06,
|
|
"loss": 0.6443942785263062,
|
|
"mean_token_accuracy": 0.79178886115551,
|
|
"num_tokens": 2184177.0,
|
|
"step": 267
|
|
},
|
|
{
|
|
"entropy": 0.5196888893842697,
|
|
"epoch": 0.11722252597047567,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 2.8668261861384045e-06,
|
|
"loss": 0.5377823114395142,
|
|
"mean_token_accuracy": 0.821480929851532,
|
|
"num_tokens": 2192369.0,
|
|
"step": 268
|
|
},
|
|
{
|
|
"entropy": 0.4297257959842682,
|
|
"epoch": 0.11765992345544013,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 2.849475848838749e-06,
|
|
"loss": 0.4501199722290039,
|
|
"mean_token_accuracy": 0.8475073277950287,
|
|
"num_tokens": 2200561.0,
|
|
"step": 269
|
|
},
|
|
{
|
|
"entropy": 0.5126072689890862,
|
|
"epoch": 0.1180973209404046,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 2.83210832616042e-06,
|
|
"loss": 0.5289058089256287,
|
|
"mean_token_accuracy": 0.8217253237962723,
|
|
"num_tokens": 2208753.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 0.539338693022728,
|
|
"epoch": 0.11853471842536906,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 2.814724472146564e-06,
|
|
"loss": 0.5616109371185303,
|
|
"mean_token_accuracy": 0.8209921717643738,
|
|
"num_tokens": 2216945.0,
|
|
"step": 271
|
|
},
|
|
{
|
|
"entropy": 0.43294807523489,
|
|
"epoch": 0.11897211591033352,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 2.7973251416434176e-06,
|
|
"loss": 0.451716810464859,
|
|
"mean_token_accuracy": 0.8498289287090302,
|
|
"num_tokens": 2225137.0,
|
|
"step": 272
|
|
},
|
|
{
|
|
"entropy": 0.47712908685207367,
|
|
"epoch": 0.11940951339529798,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 2.7799111902582697e-06,
|
|
"loss": 0.5194007754325867,
|
|
"mean_token_accuracy": 0.8256353735923767,
|
|
"num_tokens": 2233329.0,
|
|
"step": 273
|
|
},
|
|
{
|
|
"entropy": 0.47110161930322647,
|
|
"epoch": 0.11984691088026243,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 2.762483474317387e-06,
|
|
"loss": 0.49832338094711304,
|
|
"mean_token_accuracy": 0.8338220864534378,
|
|
"num_tokens": 2241521.0,
|
|
"step": 274
|
|
},
|
|
{
|
|
"entropy": 0.49556952714920044,
|
|
"epoch": 0.1202843083652269,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 2.7450428508239024e-06,
|
|
"loss": 0.5383248925209045,
|
|
"mean_token_accuracy": 0.8240469396114349,
|
|
"num_tokens": 2249713.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 0.43116144835948944,
|
|
"epoch": 0.12072170585019136,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 2.7275901774156753e-06,
|
|
"loss": 0.47572997212409973,
|
|
"mean_token_accuracy": 0.847318559885025,
|
|
"num_tokens": 2257797.0,
|
|
"step": 276
|
|
},
|
|
{
|
|
"entropy": 0.5458333194255829,
|
|
"epoch": 0.12115910333515582,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 2.710126312323119e-06,
|
|
"loss": 0.5907471776008606,
|
|
"mean_token_accuracy": 0.8092619776725769,
|
|
"num_tokens": 2265989.0,
|
|
"step": 277
|
|
},
|
|
{
|
|
"entropy": 0.4849350154399872,
|
|
"epoch": 0.12159650082012029,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 2.6926521143269914e-06,
|
|
"loss": 0.5167871713638306,
|
|
"mean_token_accuracy": 0.8285679519176483,
|
|
"num_tokens": 2274181.0,
|
|
"step": 278
|
|
},
|
|
{
|
|
"entropy": 0.48804011195898056,
|
|
"epoch": 0.12203389830508475,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 2.6751684427161684e-06,
|
|
"loss": 0.5357515215873718,
|
|
"mean_token_accuracy": 0.8296676427125931,
|
|
"num_tokens": 2282373.0,
|
|
"step": 279
|
|
},
|
|
{
|
|
"entropy": 0.4341088831424713,
|
|
"epoch": 0.1224712957900492,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 2.6576761572453903e-06,
|
|
"loss": 0.461492657661438,
|
|
"mean_token_accuracy": 0.8447287380695343,
|
|
"num_tokens": 2290388.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 0.39888056367635727,
|
|
"epoch": 0.12290869327501366,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 2.6401761180929798e-06,
|
|
"loss": 0.4248638451099396,
|
|
"mean_token_accuracy": 0.8599706888198853,
|
|
"num_tokens": 2298580.0,
|
|
"step": 281
|
|
},
|
|
{
|
|
"entropy": 0.39297056198120117,
|
|
"epoch": 0.12334609075997813,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 2.6226691858185454e-06,
|
|
"loss": 0.40902504324913025,
|
|
"mean_token_accuracy": 0.8653470277786255,
|
|
"num_tokens": 2306772.0,
|
|
"step": 282
|
|
},
|
|
{
|
|
"entropy": 0.4313372001051903,
|
|
"epoch": 0.12378348824494259,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 2.6051562213206633e-06,
|
|
"loss": 0.44663628935813904,
|
|
"mean_token_accuracy": 0.8475073277950287,
|
|
"num_tokens": 2314964.0,
|
|
"step": 283
|
|
},
|
|
{
|
|
"entropy": 0.47445718944072723,
|
|
"epoch": 0.12422088572990705,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 2.587638085794541e-06,
|
|
"loss": 0.5170879364013672,
|
|
"mean_token_accuracy": 0.8340664654970169,
|
|
"num_tokens": 2323156.0,
|
|
"step": 284
|
|
},
|
|
{
|
|
"entropy": 0.44119589775800705,
|
|
"epoch": 0.12465828321487152,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 2.5701156406896726e-06,
|
|
"loss": 0.4654120206832886,
|
|
"mean_token_accuracy": 0.8454301059246063,
|
|
"num_tokens": 2331348.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 0.5091964900493622,
|
|
"epoch": 0.12509568069983598,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 2.5525897476674722e-06,
|
|
"loss": 0.5454074144363403,
|
|
"mean_token_accuracy": 0.81659334897995,
|
|
"num_tokens": 2339540.0,
|
|
"step": 286
|
|
},
|
|
{
|
|
"entropy": 0.48265135288238525,
|
|
"epoch": 0.12553307818480045,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 2.535061268558906e-06,
|
|
"loss": 0.5154399275779724,
|
|
"mean_token_accuracy": 0.8330889493227005,
|
|
"num_tokens": 2347732.0,
|
|
"step": 287
|
|
},
|
|
{
|
|
"entropy": 0.47054049372673035,
|
|
"epoch": 0.1259704756697649,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 2.5175310653221092e-06,
|
|
"loss": 0.4928542673587799,
|
|
"mean_token_accuracy": 0.8404203355312347,
|
|
"num_tokens": 2355924.0,
|
|
"step": 288
|
|
},
|
|
{
|
|
"entropy": 0.5776143297553062,
|
|
"epoch": 0.12640787315472937,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 2.5e-06,
|
|
"loss": 0.6230844259262085,
|
|
"mean_token_accuracy": 0.7988758534193039,
|
|
"num_tokens": 2364116.0,
|
|
"step": 289
|
|
},
|
|
{
|
|
"entropy": 0.5465107038617134,
|
|
"epoch": 0.1268452706396938,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 2.482468934677891e-06,
|
|
"loss": 0.5890681743621826,
|
|
"mean_token_accuracy": 0.8060850352048874,
|
|
"num_tokens": 2372308.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 0.47898172587156296,
|
|
"epoch": 0.12728266812465827,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 2.464938731441095e-06,
|
|
"loss": 0.5131503939628601,
|
|
"mean_token_accuracy": 0.8317448645830154,
|
|
"num_tokens": 2380500.0,
|
|
"step": 291
|
|
},
|
|
{
|
|
"entropy": 0.5422700941562653,
|
|
"epoch": 0.12772006560962273,
|
|
"grad_norm": 2.609375,
|
|
"learning_rate": 2.447410252332528e-06,
|
|
"loss": 0.5898529887199402,
|
|
"mean_token_accuracy": 0.8073069453239441,
|
|
"num_tokens": 2388692.0,
|
|
"step": 292
|
|
},
|
|
{
|
|
"entropy": 0.49293333292007446,
|
|
"epoch": 0.1281574630945872,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 2.429884359310328e-06,
|
|
"loss": 0.5128275156021118,
|
|
"mean_token_accuracy": 0.8294232487678528,
|
|
"num_tokens": 2396884.0,
|
|
"step": 293
|
|
},
|
|
{
|
|
"entropy": 0.435122549533844,
|
|
"epoch": 0.12859486057955166,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 2.41236191420546e-06,
|
|
"loss": 0.47749215364456177,
|
|
"mean_token_accuracy": 0.8449413478374481,
|
|
"num_tokens": 2405076.0,
|
|
"step": 294
|
|
},
|
|
{
|
|
"entropy": 0.4070725217461586,
|
|
"epoch": 0.12903225806451613,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 2.394843778679338e-06,
|
|
"loss": 0.44448477029800415,
|
|
"mean_token_accuracy": 0.8543499559164047,
|
|
"num_tokens": 2413268.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 0.4112040773034096,
|
|
"epoch": 0.1294696555494806,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 2.377330814181455e-06,
|
|
"loss": 0.4309145510196686,
|
|
"mean_token_accuracy": 0.8550830781459808,
|
|
"num_tokens": 2421460.0,
|
|
"step": 296
|
|
},
|
|
{
|
|
"entropy": 0.5157284215092659,
|
|
"epoch": 0.12990705303444505,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 2.3598238819070206e-06,
|
|
"loss": 0.5801538228988647,
|
|
"mean_token_accuracy": 0.8158602267503738,
|
|
"num_tokens": 2429652.0,
|
|
"step": 297
|
|
},
|
|
{
|
|
"entropy": 0.4528968632221222,
|
|
"epoch": 0.13034445051940952,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 2.3423238427546105e-06,
|
|
"loss": 0.4707263112068176,
|
|
"mean_token_accuracy": 0.8461632430553436,
|
|
"num_tokens": 2437844.0,
|
|
"step": 298
|
|
},
|
|
{
|
|
"entropy": 0.5282165333628654,
|
|
"epoch": 0.13078184800437398,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 2.3248315572838316e-06,
|
|
"loss": 0.5393767356872559,
|
|
"mean_token_accuracy": 0.8227028399705887,
|
|
"num_tokens": 2446036.0,
|
|
"step": 299
|
|
},
|
|
{
|
|
"entropy": 0.5359873697161674,
|
|
"epoch": 0.13121924548933844,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 2.30734788567301e-06,
|
|
"loss": 0.5740209817886353,
|
|
"mean_token_accuracy": 0.8135386258363724,
|
|
"num_tokens": 2454228.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 0.5769957453012466,
|
|
"epoch": 0.1316566429743029,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 2.2898736876768816e-06,
|
|
"loss": 0.6145251393318176,
|
|
"mean_token_accuracy": 0.8029080927371979,
|
|
"num_tokens": 2462420.0,
|
|
"step": 301
|
|
},
|
|
{
|
|
"entropy": 0.42022210359573364,
|
|
"epoch": 0.13209404045926737,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 2.272409822584325e-06,
|
|
"loss": 0.4709186851978302,
|
|
"mean_token_accuracy": 0.8477517068386078,
|
|
"num_tokens": 2470612.0,
|
|
"step": 302
|
|
},
|
|
{
|
|
"entropy": 0.5215721800923347,
|
|
"epoch": 0.13253143794423183,
|
|
"grad_norm": 2.65625,
|
|
"learning_rate": 2.2549571491760985e-06,
|
|
"loss": 0.5673218965530396,
|
|
"mean_token_accuracy": 0.81659334897995,
|
|
"num_tokens": 2478804.0,
|
|
"step": 303
|
|
},
|
|
{
|
|
"entropy": 0.45102396607398987,
|
|
"epoch": 0.13296883542919627,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 2.237516525682614e-06,
|
|
"loss": 0.48201784491539,
|
|
"mean_token_accuracy": 0.8437194675207138,
|
|
"num_tokens": 2486996.0,
|
|
"step": 304
|
|
},
|
|
{
|
|
"entropy": 0.42183125764131546,
|
|
"epoch": 0.13340623291416073,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 2.2200888097417308e-06,
|
|
"loss": 0.44949376583099365,
|
|
"mean_token_accuracy": 0.8519061654806137,
|
|
"num_tokens": 2495188.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 0.5075167715549469,
|
|
"epoch": 0.1338436303991252,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 2.2026748583565824e-06,
|
|
"loss": 0.5390216112136841,
|
|
"mean_token_accuracy": 0.8218998908996582,
|
|
"num_tokens": 2503172.0,
|
|
"step": 306
|
|
},
|
|
{
|
|
"entropy": 0.5761678591370583,
|
|
"epoch": 0.13428102788408966,
|
|
"grad_norm": 2.75,
|
|
"learning_rate": 2.1852755278534373e-06,
|
|
"loss": 0.6333814859390259,
|
|
"mean_token_accuracy": 0.7954545468091965,
|
|
"num_tokens": 2511364.0,
|
|
"step": 307
|
|
},
|
|
{
|
|
"entropy": 0.4733138978481293,
|
|
"epoch": 0.13471842536905412,
|
|
"grad_norm": 2.515625,
|
|
"learning_rate": 2.167891673839581e-06,
|
|
"loss": 0.5121445059776306,
|
|
"mean_token_accuracy": 0.8380987197160721,
|
|
"num_tokens": 2519556.0,
|
|
"step": 308
|
|
},
|
|
{
|
|
"entropy": 0.5125866606831551,
|
|
"epoch": 0.1351558228540186,
|
|
"grad_norm": 2.515625,
|
|
"learning_rate": 2.1505241511612522e-06,
|
|
"loss": 0.5592678785324097,
|
|
"mean_token_accuracy": 0.819892480969429,
|
|
"num_tokens": 2527748.0,
|
|
"step": 309
|
|
},
|
|
{
|
|
"entropy": 0.4545166566967964,
|
|
"epoch": 0.13559322033898305,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 2.133173813861596e-06,
|
|
"loss": 0.4897732734680176,
|
|
"mean_token_accuracy": 0.8401759415864944,
|
|
"num_tokens": 2535940.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 0.4526529014110565,
|
|
"epoch": 0.13603061782394751,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 2.1158415151386746e-06,
|
|
"loss": 0.48374703526496887,
|
|
"mean_token_accuracy": 0.8405425250530243,
|
|
"num_tokens": 2544132.0,
|
|
"step": 311
|
|
},
|
|
{
|
|
"entropy": 0.5908744037151337,
|
|
"epoch": 0.13646801530891198,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 2.098528107303508e-06,
|
|
"loss": 0.6271101832389832,
|
|
"mean_token_accuracy": 0.796676442027092,
|
|
"num_tokens": 2552324.0,
|
|
"step": 312
|
|
},
|
|
{
|
|
"entropy": 0.49712150543928146,
|
|
"epoch": 0.13690541279387644,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 2.0812344417381595e-06,
|
|
"loss": 0.5376288890838623,
|
|
"mean_token_accuracy": 0.8271016776561737,
|
|
"num_tokens": 2560516.0,
|
|
"step": 313
|
|
},
|
|
{
|
|
"entropy": 0.5063036307692528,
|
|
"epoch": 0.1373428102788409,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 2.0639613688538733e-06,
|
|
"loss": 0.5550069808959961,
|
|
"mean_token_accuracy": 0.8277125954627991,
|
|
"num_tokens": 2568708.0,
|
|
"step": 314
|
|
},
|
|
{
|
|
"entropy": 0.49224531650543213,
|
|
"epoch": 0.13778020776380537,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 2.0467097380492547e-06,
|
|
"loss": 0.5338283777236938,
|
|
"mean_token_accuracy": 0.826735109090805,
|
|
"num_tokens": 2576900.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 0.45498689264059067,
|
|
"epoch": 0.13821760524876983,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 2.0294803976685006e-06,
|
|
"loss": 0.4753957986831665,
|
|
"mean_token_accuracy": 0.8456840515136719,
|
|
"num_tokens": 2584810.0,
|
|
"step": 316
|
|
},
|
|
{
|
|
"entropy": 0.4325580522418022,
|
|
"epoch": 0.13865500273373427,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 2.01227419495968e-06,
|
|
"loss": 0.4696834087371826,
|
|
"mean_token_accuracy": 0.8431084901094437,
|
|
"num_tokens": 2593002.0,
|
|
"step": 317
|
|
},
|
|
{
|
|
"entropy": 0.5289476662874222,
|
|
"epoch": 0.13909240021869873,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 1.9950919760330757e-06,
|
|
"loss": 0.5645922422409058,
|
|
"mean_token_accuracy": 0.8153714537620544,
|
|
"num_tokens": 2601194.0,
|
|
"step": 318
|
|
},
|
|
{
|
|
"entropy": 0.4893813133239746,
|
|
"epoch": 0.1395297977036632,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 1.9779345858195757e-06,
|
|
"loss": 0.5062471628189087,
|
|
"mean_token_accuracy": 0.8324780017137527,
|
|
"num_tokens": 2609386.0,
|
|
"step": 319
|
|
},
|
|
{
|
|
"entropy": 0.5512858629226685,
|
|
"epoch": 0.13996719518862766,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 1.960802868029119e-06,
|
|
"loss": 0.5786765813827515,
|
|
"mean_token_accuracy": 0.8102394938468933,
|
|
"num_tokens": 2617578.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 0.45561161637306213,
|
|
"epoch": 0.14040459267359212,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 1.9436976651092143e-06,
|
|
"loss": 0.476117879152298,
|
|
"mean_token_accuracy": 0.8421309888362885,
|
|
"num_tokens": 2625770.0,
|
|
"step": 321
|
|
},
|
|
{
|
|
"entropy": 0.525338664650917,
|
|
"epoch": 0.14084199015855658,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 1.926619818203506e-06,
|
|
"loss": 0.563406229019165,
|
|
"mean_token_accuracy": 0.8158602118492126,
|
|
"num_tokens": 2633962.0,
|
|
"step": 322
|
|
},
|
|
{
|
|
"entropy": 0.49841468036174774,
|
|
"epoch": 0.14127938764352105,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 1.909570167110415e-06,
|
|
"loss": 0.5303950309753418,
|
|
"mean_token_accuracy": 0.8271016478538513,
|
|
"num_tokens": 2642154.0,
|
|
"step": 323
|
|
},
|
|
{
|
|
"entropy": 0.4283083751797676,
|
|
"epoch": 0.1417167851284855,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 1.8925495502418407e-06,
|
|
"loss": 0.457165002822876,
|
|
"mean_token_accuracy": 0.8508064597845078,
|
|
"num_tokens": 2650346.0,
|
|
"step": 324
|
|
},
|
|
{
|
|
"entropy": 0.4319509118795395,
|
|
"epoch": 0.14215418261344998,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 1.8755588045819325e-06,
|
|
"loss": 0.4763701856136322,
|
|
"mean_token_accuracy": 0.8459232449531555,
|
|
"num_tokens": 2658502.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 0.5478535592556,
|
|
"epoch": 0.14259158009841444,
|
|
"grad_norm": 2.546875,
|
|
"learning_rate": 1.8585987656459337e-06,
|
|
"loss": 0.6074569821357727,
|
|
"mean_token_accuracy": 0.8050237745046616,
|
|
"num_tokens": 2666657.0,
|
|
"step": 326
|
|
},
|
|
{
|
|
"entropy": 0.5283005684614182,
|
|
"epoch": 0.1430289775833789,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 1.841670267439088e-06,
|
|
"loss": 0.5483436584472656,
|
|
"mean_token_accuracy": 0.8168377429246902,
|
|
"num_tokens": 2674849.0,
|
|
"step": 327
|
|
},
|
|
{
|
|
"entropy": 0.43779096752405167,
|
|
"epoch": 0.14346637506834337,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 1.8247741424156373e-06,
|
|
"loss": 0.4743049740791321,
|
|
"mean_token_accuracy": 0.8424975425004959,
|
|
"num_tokens": 2683041.0,
|
|
"step": 328
|
|
},
|
|
{
|
|
"entropy": 0.5099033936858177,
|
|
"epoch": 0.14390377255330783,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 1.8079112214378769e-06,
|
|
"loss": 0.5568172931671143,
|
|
"mean_token_accuracy": 0.8189149498939514,
|
|
"num_tokens": 2691233.0,
|
|
"step": 329
|
|
},
|
|
{
|
|
"entropy": 0.4858943372964859,
|
|
"epoch": 0.1443411700382723,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 1.7910823337353062e-06,
|
|
"loss": 0.5260504484176636,
|
|
"mean_token_accuracy": 0.8301564007997513,
|
|
"num_tokens": 2699425.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 0.5257250666618347,
|
|
"epoch": 0.14477856752323673,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 1.7742883068638447e-06,
|
|
"loss": 0.5469223856925964,
|
|
"mean_token_accuracy": 0.8225806504487991,
|
|
"num_tokens": 2707617.0,
|
|
"step": 331
|
|
},
|
|
{
|
|
"entropy": 0.42773452401161194,
|
|
"epoch": 0.1452159650082012,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 1.7575299666651413e-06,
|
|
"loss": 0.4414147138595581,
|
|
"mean_token_accuracy": 0.8527614921331406,
|
|
"num_tokens": 2715809.0,
|
|
"step": 332
|
|
},
|
|
{
|
|
"entropy": 0.500542514026165,
|
|
"epoch": 0.14565336249316566,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 1.7408081372259633e-06,
|
|
"loss": 0.5075960159301758,
|
|
"mean_token_accuracy": 0.8308895230293274,
|
|
"num_tokens": 2724001.0,
|
|
"step": 333
|
|
},
|
|
{
|
|
"entropy": 0.4560134708881378,
|
|
"epoch": 0.14609075997813012,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 1.7241236408376702e-06,
|
|
"loss": 0.48843127489089966,
|
|
"mean_token_accuracy": 0.8379765450954437,
|
|
"num_tokens": 2732193.0,
|
|
"step": 334
|
|
},
|
|
{
|
|
"entropy": 0.5988139659166336,
|
|
"epoch": 0.14652815746309458,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 1.7074772979557802e-06,
|
|
"loss": 0.6490610241889954,
|
|
"mean_token_accuracy": 0.7976539582014084,
|
|
"num_tokens": 2740385.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 0.3993486016988754,
|
|
"epoch": 0.14696555494805905,
|
|
"grad_norm": 1.9765625,
|
|
"learning_rate": 1.6908699271596213e-06,
|
|
"loss": 0.42430931329727173,
|
|
"mean_token_accuracy": 0.8637585490942001,
|
|
"num_tokens": 2748577.0,
|
|
"step": 336
|
|
},
|
|
{
|
|
"entropy": 0.48627281934022903,
|
|
"epoch": 0.1474029524330235,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 1.6743023451120831e-06,
|
|
"loss": 0.5066972374916077,
|
|
"mean_token_accuracy": 0.8312561064958572,
|
|
"num_tokens": 2756769.0,
|
|
"step": 337
|
|
},
|
|
{
|
|
"entropy": 0.4686634689569473,
|
|
"epoch": 0.14784034991798797,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 1.6577753665194502e-06,
|
|
"loss": 0.4975348711013794,
|
|
"mean_token_accuracy": 0.8268572986125946,
|
|
"num_tokens": 2764961.0,
|
|
"step": 338
|
|
},
|
|
{
|
|
"entropy": 0.4240003898739815,
|
|
"epoch": 0.14827774740295244,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 1.6412898040913471e-06,
|
|
"loss": 0.44500935077667236,
|
|
"mean_token_accuracy": 0.8515395820140839,
|
|
"num_tokens": 2773153.0,
|
|
"step": 339
|
|
},
|
|
{
|
|
"entropy": 0.5179468542337418,
|
|
"epoch": 0.1487151448879169,
|
|
"grad_norm": 2.5625,
|
|
"learning_rate": 1.6248464685007692e-06,
|
|
"loss": 0.5472849607467651,
|
|
"mean_token_accuracy": 0.8209921717643738,
|
|
"num_tokens": 2781345.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 0.45416006445884705,
|
|
"epoch": 0.14915254237288136,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 1.6084461683442176e-06,
|
|
"loss": 0.47359973192214966,
|
|
"mean_token_accuracy": 0.8380987346172333,
|
|
"num_tokens": 2789537.0,
|
|
"step": 341
|
|
},
|
|
{
|
|
"entropy": 0.5557916834950447,
|
|
"epoch": 0.14958993985784583,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 1.5920897101019384e-06,
|
|
"loss": 0.586639940738678,
|
|
"mean_token_accuracy": 0.811950147151947,
|
|
"num_tokens": 2797729.0,
|
|
"step": 342
|
|
},
|
|
{
|
|
"entropy": 0.419264979660511,
|
|
"epoch": 0.1500273373428103,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 1.5757778980982627e-06,
|
|
"loss": 0.4477095305919647,
|
|
"mean_token_accuracy": 0.8466520011425018,
|
|
"num_tokens": 2805921.0,
|
|
"step": 343
|
|
},
|
|
{
|
|
"entropy": 0.4657460153102875,
|
|
"epoch": 0.15046473482777473,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 1.5595115344620575e-06,
|
|
"loss": 0.49677377939224243,
|
|
"mean_token_accuracy": 0.8355327546596527,
|
|
"num_tokens": 2814113.0,
|
|
"step": 344
|
|
},
|
|
{
|
|
"entropy": 0.42323917895555496,
|
|
"epoch": 0.1509021323127392,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 1.5432914190872757e-06,
|
|
"loss": 0.4586298167705536,
|
|
"mean_token_accuracy": 0.8498289436101913,
|
|
"num_tokens": 2822305.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 0.4431135877966881,
|
|
"epoch": 0.15133952979770365,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 1.5271183495936273e-06,
|
|
"loss": 0.49553415179252625,
|
|
"mean_token_accuracy": 0.8379765301942825,
|
|
"num_tokens": 2830497.0,
|
|
"step": 346
|
|
},
|
|
{
|
|
"entropy": 0.5104446560144424,
|
|
"epoch": 0.15177692728266812,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 1.5109931212873535e-06,
|
|
"loss": 0.5375692844390869,
|
|
"mean_token_accuracy": 0.8218475133180618,
|
|
"num_tokens": 2838689.0,
|
|
"step": 347
|
|
},
|
|
{
|
|
"entropy": 0.5319401547312737,
|
|
"epoch": 0.15221432476763258,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 1.494916527122116e-06,
|
|
"loss": 0.587618887424469,
|
|
"mean_token_accuracy": 0.8161045908927917,
|
|
"num_tokens": 2846881.0,
|
|
"step": 348
|
|
},
|
|
{
|
|
"entropy": 0.5874071270227432,
|
|
"epoch": 0.15265172225259704,
|
|
"grad_norm": 2.71875,
|
|
"learning_rate": 1.47888935766001e-06,
|
|
"loss": 0.6408997774124146,
|
|
"mean_token_accuracy": 0.7987536638975143,
|
|
"num_tokens": 2855073.0,
|
|
"step": 349
|
|
},
|
|
{
|
|
"entropy": 0.5707217156887054,
|
|
"epoch": 0.1530891197375615,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 1.4629124010326808e-06,
|
|
"loss": 0.6072824001312256,
|
|
"mean_token_accuracy": 0.8013196438550949,
|
|
"num_tokens": 2863265.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 0.47589151561260223,
|
|
"epoch": 0.15352651722252597,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 1.446986442902574e-06,
|
|
"loss": 0.5004896521568298,
|
|
"mean_token_accuracy": 0.8291788846254349,
|
|
"num_tokens": 2871457.0,
|
|
"step": 351
|
|
},
|
|
{
|
|
"entropy": 0.43671607226133347,
|
|
"epoch": 0.15396391470749043,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 1.4311122664242955e-06,
|
|
"loss": 0.46961554884910583,
|
|
"mean_token_accuracy": 0.8429863154888153,
|
|
"num_tokens": 2879649.0,
|
|
"step": 352
|
|
},
|
|
{
|
|
"entropy": 0.4778784140944481,
|
|
"epoch": 0.1544013121924549,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 1.415290652206105e-06,
|
|
"loss": 0.49795201420783997,
|
|
"mean_token_accuracy": 0.834921807050705,
|
|
"num_tokens": 2887841.0,
|
|
"step": 353
|
|
},
|
|
{
|
|
"entropy": 0.4867304265499115,
|
|
"epoch": 0.15483870967741936,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 1.399522378271527e-06,
|
|
"loss": 0.5051119327545166,
|
|
"mean_token_accuracy": 0.83321113884449,
|
|
"num_tokens": 2896033.0,
|
|
"step": 354
|
|
},
|
|
{
|
|
"entropy": 0.5506188422441483,
|
|
"epoch": 0.15527610716238383,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 1.3838082200210932e-06,
|
|
"loss": 0.5850009322166443,
|
|
"mean_token_accuracy": 0.8002199530601501,
|
|
"num_tokens": 2904225.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 0.5183043852448463,
|
|
"epoch": 0.1557135046473483,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 1.3681489501942077e-06,
|
|
"loss": 0.5410799980163574,
|
|
"mean_token_accuracy": 0.82233627140522,
|
|
"num_tokens": 2912417.0,
|
|
"step": 356
|
|
},
|
|
{
|
|
"entropy": 0.5144079178571701,
|
|
"epoch": 0.15615090213231275,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 1.3525453388311554e-06,
|
|
"loss": 0.5360651612281799,
|
|
"mean_token_accuracy": 0.8218474984169006,
|
|
"num_tokens": 2920609.0,
|
|
"step": 357
|
|
},
|
|
{
|
|
"entropy": 0.5170230269432068,
|
|
"epoch": 0.1565882996172772,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 1.3369981532352317e-06,
|
|
"loss": 0.5590443015098572,
|
|
"mean_token_accuracy": 0.8179374486207962,
|
|
"num_tokens": 2928801.0,
|
|
"step": 358
|
|
},
|
|
{
|
|
"entropy": 0.4852587506175041,
|
|
"epoch": 0.15702569710224165,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 1.3215081579350058e-06,
|
|
"loss": 0.5135608315467834,
|
|
"mean_token_accuracy": 0.8256353884935379,
|
|
"num_tokens": 2936993.0,
|
|
"step": 359
|
|
},
|
|
{
|
|
"entropy": 0.5194334834814072,
|
|
"epoch": 0.15746309458720611,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 1.306076114646735e-06,
|
|
"loss": 0.5491058826446533,
|
|
"mean_token_accuracy": 0.8167155385017395,
|
|
"num_tokens": 2945185.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 0.5145406797528267,
|
|
"epoch": 0.15790049207217058,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 1.2907027822369006e-06,
|
|
"loss": 0.5343478322029114,
|
|
"mean_token_accuracy": 0.8262463212013245,
|
|
"num_tokens": 2953377.0,
|
|
"step": 361
|
|
},
|
|
{
|
|
"entropy": 0.43967072665691376,
|
|
"epoch": 0.15833788955713504,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 1.2753889166848909e-06,
|
|
"loss": 0.4568479061126709,
|
|
"mean_token_accuracy": 0.8449413329362869,
|
|
"num_tokens": 2961569.0,
|
|
"step": 362
|
|
},
|
|
{
|
|
"entropy": 0.462729349732399,
|
|
"epoch": 0.1587752870420995,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 1.2601352710458314e-06,
|
|
"loss": 0.4606771469116211,
|
|
"mean_token_accuracy": 0.8440860211849213,
|
|
"num_tokens": 2969761.0,
|
|
"step": 363
|
|
},
|
|
{
|
|
"entropy": 0.46076279133558273,
|
|
"epoch": 0.15921268452706397,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 1.2449425954135452e-06,
|
|
"loss": 0.5077764987945557,
|
|
"mean_token_accuracy": 0.8333333283662796,
|
|
"num_tokens": 2977953.0,
|
|
"step": 364
|
|
},
|
|
{
|
|
"entropy": 0.4890613555908203,
|
|
"epoch": 0.15965008201202843,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 1.2298116368836772e-06,
|
|
"loss": 0.5199705958366394,
|
|
"mean_token_accuracy": 0.8293010741472244,
|
|
"num_tokens": 2986145.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 0.5147231072187424,
|
|
"epoch": 0.1600874794969929,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 1.214743139516946e-06,
|
|
"loss": 0.547106921672821,
|
|
"mean_token_accuracy": 0.8238025456666946,
|
|
"num_tokens": 2994337.0,
|
|
"step": 366
|
|
},
|
|
{
|
|
"entropy": 0.4986709952354431,
|
|
"epoch": 0.16052487698195736,
|
|
"grad_norm": 2.53125,
|
|
"learning_rate": 1.1997378443025633e-06,
|
|
"loss": 0.5374823212623596,
|
|
"mean_token_accuracy": 0.8257575631141663,
|
|
"num_tokens": 3002529.0,
|
|
"step": 367
|
|
},
|
|
{
|
|
"entropy": 0.5491506159305573,
|
|
"epoch": 0.16096227446692182,
|
|
"grad_norm": 2.5625,
|
|
"learning_rate": 1.1847964891217923e-06,
|
|
"loss": 0.5859638452529907,
|
|
"mean_token_accuracy": 0.8053518980741501,
|
|
"num_tokens": 3010721.0,
|
|
"step": 368
|
|
},
|
|
{
|
|
"entropy": 0.5448054671287537,
|
|
"epoch": 0.1613996719518863,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 1.169919808711659e-06,
|
|
"loss": 0.5831042528152466,
|
|
"mean_token_accuracy": 0.8087732344865799,
|
|
"num_tokens": 3018913.0,
|
|
"step": 369
|
|
},
|
|
{
|
|
"entropy": 0.44894183427095413,
|
|
"epoch": 0.16183706943685075,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 1.1551085346288296e-06,
|
|
"loss": 0.47109442949295044,
|
|
"mean_token_accuracy": 0.8432306945323944,
|
|
"num_tokens": 3027105.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 0.4341334030032158,
|
|
"epoch": 0.16227446692181519,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 1.140363395213629e-06,
|
|
"loss": 0.45891037583351135,
|
|
"mean_token_accuracy": 0.8500733077526093,
|
|
"num_tokens": 3035297.0,
|
|
"step": 371
|
|
},
|
|
{
|
|
"entropy": 0.46006736159324646,
|
|
"epoch": 0.16271186440677965,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 1.1256851155542297e-06,
|
|
"loss": 0.4995250403881073,
|
|
"mean_token_accuracy": 0.8370302766561508,
|
|
"num_tokens": 3043386.0,
|
|
"step": 372
|
|
},
|
|
{
|
|
"entropy": 0.5512775182723999,
|
|
"epoch": 0.1631492618917441,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 1.1110744174509952e-06,
|
|
"loss": 0.5817916989326477,
|
|
"mean_token_accuracy": 0.8120723366737366,
|
|
"num_tokens": 3051578.0,
|
|
"step": 373
|
|
},
|
|
{
|
|
"entropy": 0.5470651388168335,
|
|
"epoch": 0.16358665937670858,
|
|
"grad_norm": 2.703125,
|
|
"learning_rate": 1.0965320193809808e-06,
|
|
"loss": 0.5833251476287842,
|
|
"mean_token_accuracy": 0.8063294142484665,
|
|
"num_tokens": 3059770.0,
|
|
"step": 374
|
|
},
|
|
{
|
|
"entropy": 0.49073220789432526,
|
|
"epoch": 0.16402405686167304,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 1.0820586364626103e-06,
|
|
"loss": 0.5221870541572571,
|
|
"mean_token_accuracy": 0.8249022364616394,
|
|
"num_tokens": 3067962.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 0.4238805025815964,
|
|
"epoch": 0.1644614543466375,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 1.0676549804205048e-06,
|
|
"loss": 0.45591771602630615,
|
|
"mean_token_accuracy": 0.8479961007833481,
|
|
"num_tokens": 3076154.0,
|
|
"step": 376
|
|
},
|
|
{
|
|
"entropy": 0.549527570605278,
|
|
"epoch": 0.16489885183160197,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 1.0533217595504859e-06,
|
|
"loss": 0.5825474858283997,
|
|
"mean_token_accuracy": 0.8071847558021545,
|
|
"num_tokens": 3084346.0,
|
|
"step": 377
|
|
},
|
|
{
|
|
"entropy": 0.4517410472035408,
|
|
"epoch": 0.16533624931656643,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 1.0390596786847435e-06,
|
|
"loss": 0.48170745372772217,
|
|
"mean_token_accuracy": 0.8393206298351288,
|
|
"num_tokens": 3092538.0,
|
|
"step": 378
|
|
},
|
|
{
|
|
"entropy": 0.4634730815887451,
|
|
"epoch": 0.1657736468015309,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 1.02486943915718e-06,
|
|
"loss": 0.5015561580657959,
|
|
"mean_token_accuracy": 0.8304007798433304,
|
|
"num_tokens": 3100730.0,
|
|
"step": 379
|
|
},
|
|
{
|
|
"entropy": 0.5204268097877502,
|
|
"epoch": 0.16621104428649536,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 1.0107517387689168e-06,
|
|
"loss": 0.5475565791130066,
|
|
"mean_token_accuracy": 0.8212365508079529,
|
|
"num_tokens": 3108922.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 0.4305594936013222,
|
|
"epoch": 0.16664844177145982,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 9.967072717539852e-07,
|
|
"loss": 0.46948927640914917,
|
|
"mean_token_accuracy": 0.8459188789129257,
|
|
"num_tokens": 3117114.0,
|
|
"step": 381
|
|
},
|
|
{
|
|
"entropy": 0.4303949624300003,
|
|
"epoch": 0.16708583925642428,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 9.82736728745186e-07,
|
|
"loss": 0.4573507308959961,
|
|
"mean_token_accuracy": 0.8488514274358749,
|
|
"num_tokens": 3125306.0,
|
|
"step": 382
|
|
},
|
|
{
|
|
"entropy": 0.5324750170111656,
|
|
"epoch": 0.16752323674138875,
|
|
"grad_norm": 2.53125,
|
|
"learning_rate": 9.688407967401248e-07,
|
|
"loss": 0.570069432258606,
|
|
"mean_token_accuracy": 0.8161045908927917,
|
|
"num_tokens": 3133498.0,
|
|
"step": 383
|
|
},
|
|
{
|
|
"entropy": 0.4431682378053665,
|
|
"epoch": 0.1679606342263532,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 9.550201590674343e-07,
|
|
"loss": 0.46429842710494995,
|
|
"mean_token_accuracy": 0.8443304151296616,
|
|
"num_tokens": 3141690.0,
|
|
"step": 384
|
|
},
|
|
{
|
|
"entropy": 0.3733269274234772,
|
|
"epoch": 0.16839803171131765,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 9.412754953531664e-07,
|
|
"loss": 0.4124542474746704,
|
|
"mean_token_accuracy": 0.8643695116043091,
|
|
"num_tokens": 3149882.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 0.42850935459136963,
|
|
"epoch": 0.1688354291962821,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 9.276074814873778e-07,
|
|
"loss": 0.47506606578826904,
|
|
"mean_token_accuracy": 0.8418866097927094,
|
|
"num_tokens": 3158074.0,
|
|
"step": 386
|
|
},
|
|
{
|
|
"entropy": 0.5115486085414886,
|
|
"epoch": 0.16927282668124657,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 9.140167895908867e-07,
|
|
"loss": 0.5542680025100708,
|
|
"mean_token_accuracy": 0.8277126103639603,
|
|
"num_tokens": 3166266.0,
|
|
"step": 387
|
|
},
|
|
{
|
|
"entropy": 0.5119223147630692,
|
|
"epoch": 0.16971022416621104,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 9.005040879822269e-07,
|
|
"loss": 0.5519320368766785,
|
|
"mean_token_accuracy": 0.8231915831565857,
|
|
"num_tokens": 3174458.0,
|
|
"step": 388
|
|
},
|
|
{
|
|
"entropy": 0.5366015583276749,
|
|
"epoch": 0.1701476216511755,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 8.870700411447817e-07,
|
|
"loss": 0.6139038801193237,
|
|
"mean_token_accuracy": 0.8113391995429993,
|
|
"num_tokens": 3182650.0,
|
|
"step": 389
|
|
},
|
|
{
|
|
"entropy": 0.45184676349163055,
|
|
"epoch": 0.17058501913613996,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 8.737153096941045e-07,
|
|
"loss": 0.46673744916915894,
|
|
"mean_token_accuracy": 0.8447262048721313,
|
|
"num_tokens": 3190815.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 0.42296013981103897,
|
|
"epoch": 0.17102241662110443,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 8.604405503454399e-07,
|
|
"loss": 0.4525824189186096,
|
|
"mean_token_accuracy": 0.8526447266340256,
|
|
"num_tokens": 3198964.0,
|
|
"step": 391
|
|
},
|
|
{
|
|
"entropy": 0.5525547116994858,
|
|
"epoch": 0.1714598141060689,
|
|
"grad_norm": 2.625,
|
|
"learning_rate": 8.472464158814256e-07,
|
|
"loss": 0.5755280256271362,
|
|
"mean_token_accuracy": 0.8087732195854187,
|
|
"num_tokens": 3207156.0,
|
|
"step": 392
|
|
},
|
|
{
|
|
"entropy": 0.4999287575483322,
|
|
"epoch": 0.17189721159103336,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 8.341335551199903e-07,
|
|
"loss": 0.518805980682373,
|
|
"mean_token_accuracy": 0.8279569894075394,
|
|
"num_tokens": 3215348.0,
|
|
"step": 393
|
|
},
|
|
{
|
|
"entropy": 0.5184216350317001,
|
|
"epoch": 0.17233460907599782,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 8.21102612882454e-07,
|
|
"loss": 0.562769889831543,
|
|
"mean_token_accuracy": 0.8147605061531067,
|
|
"num_tokens": 3223540.0,
|
|
"step": 394
|
|
},
|
|
{
|
|
"entropy": 0.4349280297756195,
|
|
"epoch": 0.17277200656096228,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 8.081542299618139e-07,
|
|
"loss": 0.45633161067962646,
|
|
"mean_token_accuracy": 0.8467741906642914,
|
|
"num_tokens": 3231732.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 0.5333330482244492,
|
|
"epoch": 0.17320940404592675,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 7.952890430912374e-07,
|
|
"loss": 0.5744578838348389,
|
|
"mean_token_accuracy": 0.8159824013710022,
|
|
"num_tokens": 3239924.0,
|
|
"step": 396
|
|
},
|
|
{
|
|
"entropy": 0.5034205839037895,
|
|
"epoch": 0.1736468015308912,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 7.825076849127458e-07,
|
|
"loss": 0.5307492017745972,
|
|
"mean_token_accuracy": 0.8242913037538528,
|
|
"num_tokens": 3248116.0,
|
|
"step": 397
|
|
},
|
|
{
|
|
"entropy": 0.471238374710083,
|
|
"epoch": 0.17408419901585565,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 7.698107839461111e-07,
|
|
"loss": 0.5049214363098145,
|
|
"mean_token_accuracy": 0.8361437022686005,
|
|
"num_tokens": 3256308.0,
|
|
"step": 398
|
|
},
|
|
{
|
|
"entropy": 0.4965353161096573,
|
|
"epoch": 0.1745215965008201,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 7.57198964557942e-07,
|
|
"loss": 0.49766212701797485,
|
|
"mean_token_accuracy": 0.8317448496818542,
|
|
"num_tokens": 3264500.0,
|
|
"step": 399
|
|
},
|
|
{
|
|
"entropy": 0.5184244439005852,
|
|
"epoch": 0.17495899398578457,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 7.446728469309872e-07,
|
|
"loss": 0.5625096559524536,
|
|
"mean_token_accuracy": 0.815004900097847,
|
|
"num_tokens": 3272692.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 0.4760790169239044,
|
|
"epoch": 0.17539639147074904,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 7.322330470336314e-07,
|
|
"loss": 0.5208472013473511,
|
|
"mean_token_accuracy": 0.8274995982646942,
|
|
"num_tokens": 3280738.0,
|
|
"step": 401
|
|
},
|
|
{
|
|
"entropy": 0.5227344185113907,
|
|
"epoch": 0.1758337889557135,
|
|
"grad_norm": 2.546875,
|
|
"learning_rate": 7.198801765896115e-07,
|
|
"loss": 0.5777583122253418,
|
|
"mean_token_accuracy": 0.8146696388721466,
|
|
"num_tokens": 3288788.0,
|
|
"step": 402
|
|
},
|
|
{
|
|
"entropy": 0.5049602687358856,
|
|
"epoch": 0.17627118644067796,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 7.076148430479321e-07,
|
|
"loss": 0.5320097804069519,
|
|
"mean_token_accuracy": 0.8231915980577469,
|
|
"num_tokens": 3296980.0,
|
|
"step": 403
|
|
},
|
|
{
|
|
"entropy": 0.4577829986810684,
|
|
"epoch": 0.17670858392564243,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 6.954376495529927e-07,
|
|
"loss": 0.49313080310821533,
|
|
"mean_token_accuracy": 0.8317448645830154,
|
|
"num_tokens": 3305172.0,
|
|
"step": 404
|
|
},
|
|
{
|
|
"entropy": 0.4651181995868683,
|
|
"epoch": 0.1771459814106069,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 6.833491949149329e-07,
|
|
"loss": 0.49456295371055603,
|
|
"mean_token_accuracy": 0.8361437022686005,
|
|
"num_tokens": 3313364.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 0.541761614382267,
|
|
"epoch": 0.17758337889557135,
|
|
"grad_norm": 2.515625,
|
|
"learning_rate": 6.713500735801811e-07,
|
|
"loss": 0.5926611423492432,
|
|
"mean_token_accuracy": 0.80840665102005,
|
|
"num_tokens": 3321556.0,
|
|
"step": 406
|
|
},
|
|
{
|
|
"entropy": 0.5580184161663055,
|
|
"epoch": 0.17802077638053582,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 6.594408756022272e-07,
|
|
"loss": 0.584408164024353,
|
|
"mean_token_accuracy": 0.8082844614982605,
|
|
"num_tokens": 3329748.0,
|
|
"step": 407
|
|
},
|
|
{
|
|
"entropy": 0.5339596942067146,
|
|
"epoch": 0.17845817386550028,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 6.476221866126029e-07,
|
|
"loss": 0.5697808265686035,
|
|
"mean_token_accuracy": 0.8195258975028992,
|
|
"num_tokens": 3337940.0,
|
|
"step": 408
|
|
},
|
|
{
|
|
"entropy": 0.43793395161628723,
|
|
"epoch": 0.17889557135046474,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 6.358945877920861e-07,
|
|
"loss": 0.46238237619400024,
|
|
"mean_token_accuracy": 0.8409090787172318,
|
|
"num_tokens": 3346132.0,
|
|
"step": 409
|
|
},
|
|
{
|
|
"entropy": 0.4602765664458275,
|
|
"epoch": 0.1793329688354292,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 6.242586558421216e-07,
|
|
"loss": 0.4967961013317108,
|
|
"mean_token_accuracy": 0.8357963114976883,
|
|
"num_tokens": 3354067.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 0.47321972250938416,
|
|
"epoch": 0.17977036632039367,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 6.127149629564605e-07,
|
|
"loss": 0.5182679295539856,
|
|
"mean_token_accuracy": 0.8352883756160736,
|
|
"num_tokens": 3362259.0,
|
|
"step": 411
|
|
},
|
|
{
|
|
"entropy": 0.4488251060247421,
|
|
"epoch": 0.1802077638053581,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 6.012640767930217e-07,
|
|
"loss": 0.5010267496109009,
|
|
"mean_token_accuracy": 0.8410312682390213,
|
|
"num_tokens": 3370451.0,
|
|
"step": 412
|
|
},
|
|
{
|
|
"entropy": 0.5726240426301956,
|
|
"epoch": 0.18064516129032257,
|
|
"grad_norm": 2.6875,
|
|
"learning_rate": 5.899065604459814e-07,
|
|
"loss": 0.630211353302002,
|
|
"mean_token_accuracy": 0.8007087111473083,
|
|
"num_tokens": 3378643.0,
|
|
"step": 413
|
|
},
|
|
{
|
|
"entropy": 0.475062720477581,
|
|
"epoch": 0.18108255877528703,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 5.786429724180801e-07,
|
|
"loss": 0.48651978373527527,
|
|
"mean_token_accuracy": 0.8344330340623856,
|
|
"num_tokens": 3386835.0,
|
|
"step": 414
|
|
},
|
|
{
|
|
"entropy": 0.5285491198301315,
|
|
"epoch": 0.1815199562602515,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 5.674738665931575e-07,
|
|
"loss": 0.5599380731582642,
|
|
"mean_token_accuracy": 0.8178152590990067,
|
|
"num_tokens": 3395027.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 0.4876674935221672,
|
|
"epoch": 0.18195735374521596,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 5.563997922089193e-07,
|
|
"loss": 0.515315055847168,
|
|
"mean_token_accuracy": 0.8255131989717484,
|
|
"num_tokens": 3403219.0,
|
|
"step": 416
|
|
},
|
|
{
|
|
"entropy": 0.4463843032717705,
|
|
"epoch": 0.18239475123018042,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 5.454212938299256e-07,
|
|
"loss": 0.4906759262084961,
|
|
"mean_token_accuracy": 0.8361436873674393,
|
|
"num_tokens": 3411411.0,
|
|
"step": 417
|
|
},
|
|
{
|
|
"entropy": 0.4502919465303421,
|
|
"epoch": 0.1828321487151449,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 5.34538911320811e-07,
|
|
"loss": 0.4786447584629059,
|
|
"mean_token_accuracy": 0.8421309888362885,
|
|
"num_tokens": 3419603.0,
|
|
"step": 418
|
|
},
|
|
{
|
|
"entropy": 0.5915882289409637,
|
|
"epoch": 0.18326954620010935,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 5.237531798197415e-07,
|
|
"loss": 0.6288676857948303,
|
|
"mean_token_accuracy": 0.8032746911048889,
|
|
"num_tokens": 3427795.0,
|
|
"step": 419
|
|
},
|
|
{
|
|
"entropy": 0.5324423611164093,
|
|
"epoch": 0.18370694368507381,
|
|
"grad_norm": 2.6875,
|
|
"learning_rate": 5.130646297120936e-07,
|
|
"loss": 0.5872654318809509,
|
|
"mean_token_accuracy": 0.8080400824546814,
|
|
"num_tokens": 3435987.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 0.5749758183956146,
|
|
"epoch": 0.18414434117003828,
|
|
"grad_norm": 2.59375,
|
|
"learning_rate": 5.02473786604378e-07,
|
|
"loss": 0.6213216185569763,
|
|
"mean_token_accuracy": 0.7925220131874084,
|
|
"num_tokens": 3444179.0,
|
|
"step": 421
|
|
},
|
|
{
|
|
"entropy": 0.41174645721912384,
|
|
"epoch": 0.18458173865500274,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 4.91981171298388e-07,
|
|
"loss": 0.4298594892024994,
|
|
"mean_token_accuracy": 0.8552329391241074,
|
|
"num_tokens": 3452181.0,
|
|
"step": 422
|
|
},
|
|
{
|
|
"entropy": 0.47275856137275696,
|
|
"epoch": 0.1850191361399672,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 4.815872997655941e-07,
|
|
"loss": 0.5014662742614746,
|
|
"mean_token_accuracy": 0.836754634976387,
|
|
"num_tokens": 3460373.0,
|
|
"step": 423
|
|
},
|
|
{
|
|
"entropy": 0.4867643192410469,
|
|
"epoch": 0.18545653362493167,
|
|
"grad_norm": 2.53125,
|
|
"learning_rate": 4.712926831217685e-07,
|
|
"loss": 0.5217025279998779,
|
|
"mean_token_accuracy": 0.8223951309919357,
|
|
"num_tokens": 3468250.0,
|
|
"step": 424
|
|
},
|
|
{
|
|
"entropy": 0.3914794325828552,
|
|
"epoch": 0.1858939311098961,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 4.610978276018496e-07,
|
|
"loss": 0.4261934161186218,
|
|
"mean_token_accuracy": 0.8550831079483032,
|
|
"num_tokens": 3476442.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 0.4488445296883583,
|
|
"epoch": 0.18633132859486057,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 4.5100323453505344e-07,
|
|
"loss": 0.4937145709991455,
|
|
"mean_token_accuracy": 0.8369990140199661,
|
|
"num_tokens": 3484634.0,
|
|
"step": 426
|
|
},
|
|
{
|
|
"entropy": 0.5070808678865433,
|
|
"epoch": 0.18676872607982503,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 4.410094003202134e-07,
|
|
"loss": 0.5438359379768372,
|
|
"mean_token_accuracy": 0.8164711743593216,
|
|
"num_tokens": 3492826.0,
|
|
"step": 427
|
|
},
|
|
{
|
|
"entropy": 0.5789096802473068,
|
|
"epoch": 0.1872061235647895,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 4.3111681640137763e-07,
|
|
"loss": 0.610559344291687,
|
|
"mean_token_accuracy": 0.8032746911048889,
|
|
"num_tokens": 3501018.0,
|
|
"step": 428
|
|
},
|
|
{
|
|
"entropy": 0.48394788056612015,
|
|
"epoch": 0.18764352104975396,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 4.2132596924363666e-07,
|
|
"loss": 0.49888381361961365,
|
|
"mean_token_accuracy": 0.8296676576137543,
|
|
"num_tokens": 3509210.0,
|
|
"step": 429
|
|
},
|
|
{
|
|
"entropy": 0.5190452337265015,
|
|
"epoch": 0.18808091853471842,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 4.116373403092058e-07,
|
|
"loss": 0.5448641777038574,
|
|
"mean_token_accuracy": 0.8240469247102737,
|
|
"num_tokens": 3517402.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 0.4755612760782242,
|
|
"epoch": 0.18851831601968289,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 4.020514060337447e-07,
|
|
"loss": 0.498245507478714,
|
|
"mean_token_accuracy": 0.8312561064958572,
|
|
"num_tokens": 3525594.0,
|
|
"step": 431
|
|
},
|
|
{
|
|
"entropy": 0.45102784037590027,
|
|
"epoch": 0.18895571350464735,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 3.925686378029331e-07,
|
|
"loss": 0.4863170087337494,
|
|
"mean_token_accuracy": 0.8376099616289139,
|
|
"num_tokens": 3533786.0,
|
|
"step": 432
|
|
},
|
|
{
|
|
"entropy": 0.4501868635416031,
|
|
"epoch": 0.1893931109896118,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 3.831895019292897e-07,
|
|
"loss": 0.47062358260154724,
|
|
"mean_token_accuracy": 0.8438416421413422,
|
|
"num_tokens": 3541978.0,
|
|
"step": 433
|
|
},
|
|
{
|
|
"entropy": 0.45614657551050186,
|
|
"epoch": 0.18983050847457628,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 3.7391445962923824e-07,
|
|
"loss": 0.48604702949523926,
|
|
"mean_token_accuracy": 0.8367546498775482,
|
|
"num_tokens": 3550170.0,
|
|
"step": 434
|
|
},
|
|
{
|
|
"entropy": 0.4634532630443573,
|
|
"epoch": 0.19026790595954074,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 3.647439670004316e-07,
|
|
"loss": 0.4933796525001526,
|
|
"mean_token_accuracy": 0.8395650088787079,
|
|
"num_tokens": 3558362.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 0.4207415208220482,
|
|
"epoch": 0.1907053034445052,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 3.5567847499932004e-07,
|
|
"loss": 0.4387015104293823,
|
|
"mean_token_accuracy": 0.8525171130895615,
|
|
"num_tokens": 3566554.0,
|
|
"step": 436
|
|
},
|
|
{
|
|
"entropy": 0.44867319613695145,
|
|
"epoch": 0.19114270092946967,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 3.4671842941897764e-07,
|
|
"loss": 0.4772151708602905,
|
|
"mean_token_accuracy": 0.8452994525432587,
|
|
"num_tokens": 3574373.0,
|
|
"step": 437
|
|
},
|
|
{
|
|
"entropy": 0.4577394872903824,
|
|
"epoch": 0.19158009841443413,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 3.3786427086718006e-07,
|
|
"loss": 0.4948572516441345,
|
|
"mean_token_accuracy": 0.8396871834993362,
|
|
"num_tokens": 3582565.0,
|
|
"step": 438
|
|
},
|
|
{
|
|
"entropy": 0.5138989388942719,
|
|
"epoch": 0.19201749589939857,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 3.2911643474473647e-07,
|
|
"loss": 0.5472657680511475,
|
|
"mean_token_accuracy": 0.8216031342744827,
|
|
"num_tokens": 3590757.0,
|
|
"step": 439
|
|
},
|
|
{
|
|
"entropy": 0.5489735826849937,
|
|
"epoch": 0.19245489338436303,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 3.2047535122408076e-07,
|
|
"loss": 0.5895254611968994,
|
|
"mean_token_accuracy": 0.8035190552473068,
|
|
"num_tokens": 3598949.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 0.5162252485752106,
|
|
"epoch": 0.1928922908693275,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 3.119414452281158e-07,
|
|
"loss": 0.5371385216712952,
|
|
"mean_token_accuracy": 0.8258797526359558,
|
|
"num_tokens": 3607141.0,
|
|
"step": 441
|
|
},
|
|
{
|
|
"entropy": 0.5523306131362915,
|
|
"epoch": 0.19332968835429196,
|
|
"grad_norm": 2.65625,
|
|
"learning_rate": 3.035151364093214e-07,
|
|
"loss": 0.5896844267845154,
|
|
"mean_token_accuracy": 0.807062566280365,
|
|
"num_tokens": 3615333.0,
|
|
"step": 442
|
|
},
|
|
{
|
|
"entropy": 0.4346807897090912,
|
|
"epoch": 0.19376708583925642,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 2.9519683912911267e-07,
|
|
"loss": 0.47453874349594116,
|
|
"mean_token_accuracy": 0.8470185697078705,
|
|
"num_tokens": 3623525.0,
|
|
"step": 443
|
|
},
|
|
{
|
|
"entropy": 0.4658875912427902,
|
|
"epoch": 0.19420448332422088,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 2.869869624374699e-07,
|
|
"loss": 0.4884585440158844,
|
|
"mean_token_accuracy": 0.8411309272050858,
|
|
"num_tokens": 3631330.0,
|
|
"step": 444
|
|
},
|
|
{
|
|
"entropy": 0.45280956476926804,
|
|
"epoch": 0.19464188080918535,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 2.788859100528196e-07,
|
|
"loss": 0.4797264635562897,
|
|
"mean_token_accuracy": 0.8361436873674393,
|
|
"num_tokens": 3639522.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 0.4983842372894287,
|
|
"epoch": 0.1950792782941498,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 2.708940803421825e-07,
|
|
"loss": 0.512587308883667,
|
|
"mean_token_accuracy": 0.8296676427125931,
|
|
"num_tokens": 3647714.0,
|
|
"step": 446
|
|
},
|
|
{
|
|
"entropy": 0.4361315816640854,
|
|
"epoch": 0.19551667577911427,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 2.630118663015849e-07,
|
|
"loss": 0.4708927273750305,
|
|
"mean_token_accuracy": 0.8421309888362885,
|
|
"num_tokens": 3655906.0,
|
|
"step": 447
|
|
},
|
|
{
|
|
"entropy": 0.49473853409290314,
|
|
"epoch": 0.19595407326407874,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 2.552396555367334e-07,
|
|
"loss": 0.5099279880523682,
|
|
"mean_token_accuracy": 0.8274682313203812,
|
|
"num_tokens": 3664098.0,
|
|
"step": 448
|
|
},
|
|
{
|
|
"entropy": 0.4408658742904663,
|
|
"epoch": 0.1963914707490432,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 2.4757783024395244e-07,
|
|
"loss": 0.46858102083206177,
|
|
"mean_token_accuracy": 0.8464076220989227,
|
|
"num_tokens": 3672290.0,
|
|
"step": 449
|
|
},
|
|
{
|
|
"entropy": 0.48043954372406006,
|
|
"epoch": 0.19682886823400766,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 2.400267671913917e-07,
|
|
"loss": 0.5275200605392456,
|
|
"mean_token_accuracy": 0.8262463361024857,
|
|
"num_tokens": 3680482.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 0.5617353320121765,
|
|
"epoch": 0.19726626571897213,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 2.325868377004986e-07,
|
|
"loss": 0.5783557891845703,
|
|
"mean_token_accuracy": 0.8124389052391052,
|
|
"num_tokens": 3688674.0,
|
|
"step": 451
|
|
},
|
|
{
|
|
"entropy": 0.4827708601951599,
|
|
"epoch": 0.19770366320393656,
|
|
"grad_norm": 2.515625,
|
|
"learning_rate": 2.2525840762775863e-07,
|
|
"loss": 0.5348269939422607,
|
|
"mean_token_accuracy": 0.8273460566997528,
|
|
"num_tokens": 3696866.0,
|
|
"step": 452
|
|
},
|
|
{
|
|
"entropy": 0.5675117522478104,
|
|
"epoch": 0.19814106068890103,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 2.1804183734670277e-07,
|
|
"loss": 0.593768835067749,
|
|
"mean_token_accuracy": 0.8073069453239441,
|
|
"num_tokens": 3705058.0,
|
|
"step": 453
|
|
},
|
|
{
|
|
"entropy": 0.5070274025201797,
|
|
"epoch": 0.1985784581738655,
|
|
"grad_norm": 2.609375,
|
|
"learning_rate": 2.1093748173018846e-07,
|
|
"loss": 0.5375203490257263,
|
|
"mean_token_accuracy": 0.8241690993309021,
|
|
"num_tokens": 3713250.0,
|
|
"step": 454
|
|
},
|
|
{
|
|
"entropy": 0.457052618265152,
|
|
"epoch": 0.19901585565882995,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 2.039456901329473e-07,
|
|
"loss": 0.5010364651679993,
|
|
"mean_token_accuracy": 0.8365102708339691,
|
|
"num_tokens": 3721442.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 0.5044142678380013,
|
|
"epoch": 0.19945325314379442,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 1.9706680637440745e-07,
|
|
"loss": 0.511577844619751,
|
|
"mean_token_accuracy": 0.8309101909399033,
|
|
"num_tokens": 3729588.0,
|
|
"step": 456
|
|
},
|
|
{
|
|
"entropy": 0.5089931935071945,
|
|
"epoch": 0.19989065062875888,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 1.9030116872178317e-07,
|
|
"loss": 0.5382363796234131,
|
|
"mean_token_accuracy": 0.8235581666231155,
|
|
"num_tokens": 3737780.0,
|
|
"step": 457
|
|
},
|
|
{
|
|
"entropy": 0.45246878266334534,
|
|
"epoch": 0.20032804811372334,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 1.836491098734447e-07,
|
|
"loss": 0.4766382873058319,
|
|
"mean_token_accuracy": 0.8478040844202042,
|
|
"num_tokens": 3745662.0,
|
|
"step": 458
|
|
},
|
|
{
|
|
"entropy": 0.49909115582704544,
|
|
"epoch": 0.2007654455986878,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 1.771109569425547e-07,
|
|
"loss": 0.532217800617218,
|
|
"mean_token_accuracy": 0.8260019570589066,
|
|
"num_tokens": 3753854.0,
|
|
"step": 459
|
|
},
|
|
{
|
|
"entropy": 0.46753478050231934,
|
|
"epoch": 0.20120284308365227,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 1.7068703144098365e-07,
|
|
"loss": 0.49318355321884155,
|
|
"mean_token_accuracy": 0.8355327546596527,
|
|
"num_tokens": 3762046.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 0.4208934009075165,
|
|
"epoch": 0.20164024056861674,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 1.6437764926350074e-07,
|
|
"loss": 0.4540482759475708,
|
|
"mean_token_accuracy": 0.8530058562755585,
|
|
"num_tokens": 3770238.0,
|
|
"step": 461
|
|
},
|
|
{
|
|
"entropy": 0.4388285279273987,
|
|
"epoch": 0.2020776380535812,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 1.5818312067223673e-07,
|
|
"loss": 0.45131394267082214,
|
|
"mean_token_accuracy": 0.8472629636526108,
|
|
"num_tokens": 3778430.0,
|
|
"step": 462
|
|
},
|
|
{
|
|
"entropy": 0.4319133758544922,
|
|
"epoch": 0.20251503553854566,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 1.5210375028143097e-07,
|
|
"loss": 0.46854692697525024,
|
|
"mean_token_accuracy": 0.8405425101518631,
|
|
"num_tokens": 3786622.0,
|
|
"step": 463
|
|
},
|
|
{
|
|
"entropy": 0.49125488102436066,
|
|
"epoch": 0.20295243302351013,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 1.4613983704244826e-07,
|
|
"loss": 0.5005199909210205,
|
|
"mean_token_accuracy": 0.8341886550188065,
|
|
"num_tokens": 3794814.0,
|
|
"step": 464
|
|
},
|
|
{
|
|
"entropy": 0.42622826248407364,
|
|
"epoch": 0.2033898305084746,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 1.4029167422908107e-07,
|
|
"loss": 0.4538479447364807,
|
|
"mean_token_accuracy": 0.8445759415626526,
|
|
"num_tokens": 3802759.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 0.5118953883647919,
|
|
"epoch": 0.20382722799343903,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 1.345595494231261e-07,
|
|
"loss": 0.5550968647003174,
|
|
"mean_token_accuracy": 0.8173265010118484,
|
|
"num_tokens": 3810951.0,
|
|
"step": 466
|
|
},
|
|
{
|
|
"entropy": 0.4239061772823334,
|
|
"epoch": 0.2042646254784035,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 1.2894374450024338e-07,
|
|
"loss": 0.43111279606819153,
|
|
"mean_token_accuracy": 0.8519061654806137,
|
|
"num_tokens": 3819143.0,
|
|
"step": 467
|
|
},
|
|
{
|
|
"entropy": 0.4749840125441551,
|
|
"epoch": 0.20470202296336795,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 1.234445356160932e-07,
|
|
"loss": 0.5093938112258911,
|
|
"mean_token_accuracy": 0.8295454531908035,
|
|
"num_tokens": 3827335.0,
|
|
"step": 468
|
|
},
|
|
{
|
|
"entropy": 0.48437564074993134,
|
|
"epoch": 0.20513942044833242,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 1.1806219319275918e-07,
|
|
"loss": 0.524286150932312,
|
|
"mean_token_accuracy": 0.8301564007997513,
|
|
"num_tokens": 3835527.0,
|
|
"step": 469
|
|
},
|
|
{
|
|
"entropy": 0.5392681062221527,
|
|
"epoch": 0.20557681793329688,
|
|
"grad_norm": 2.53125,
|
|
"learning_rate": 1.1279698190544918e-07,
|
|
"loss": 0.5733947157859802,
|
|
"mean_token_accuracy": 0.813049852848053,
|
|
"num_tokens": 3843719.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 0.38556477427482605,
|
|
"epoch": 0.20601421541826134,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 1.0764916066947795e-07,
|
|
"loss": 0.41268569231033325,
|
|
"mean_token_accuracy": 0.8622882515192032,
|
|
"num_tokens": 3851867.0,
|
|
"step": 471
|
|
},
|
|
{
|
|
"entropy": 0.5322115048766136,
|
|
"epoch": 0.2064516129032258,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 1.0261898262753811e-07,
|
|
"loss": 0.5719882249832153,
|
|
"mean_token_accuracy": 0.8161045908927917,
|
|
"num_tokens": 3860059.0,
|
|
"step": 472
|
|
},
|
|
{
|
|
"entropy": 0.45311254262924194,
|
|
"epoch": 0.20688901038819027,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 9.770669513725128e-08,
|
|
"loss": 0.4832289218902588,
|
|
"mean_token_accuracy": 0.8361436724662781,
|
|
"num_tokens": 3868251.0,
|
|
"step": 473
|
|
},
|
|
{
|
|
"entropy": 0.5146927535533905,
|
|
"epoch": 0.20732640787315473,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 9.291253975900138e-08,
|
|
"loss": 0.5571263432502747,
|
|
"mean_token_accuracy": 0.8212365657091141,
|
|
"num_tokens": 3876443.0,
|
|
"step": 474
|
|
},
|
|
{
|
|
"entropy": 0.47573499381542206,
|
|
"epoch": 0.2077638053581192,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 8.823675224406052e-08,
|
|
"loss": 0.5182182788848877,
|
|
"mean_token_accuracy": 0.8257741034030914,
|
|
"num_tokens": 3884497.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 0.5343288853764534,
|
|
"epoch": 0.20820120284308366,
|
|
"grad_norm": 2.625,
|
|
"learning_rate": 8.367956252299192e-08,
|
|
"loss": 0.5817354917526245,
|
|
"mean_token_accuracy": 0.8120723366737366,
|
|
"num_tokens": 3892689.0,
|
|
"step": 476
|
|
},
|
|
{
|
|
"entropy": 0.5038627311587334,
|
|
"epoch": 0.20863860032804812,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 7.924119469434666e-08,
|
|
"loss": 0.5361433029174805,
|
|
"mean_token_accuracy": 0.8236803561449051,
|
|
"num_tokens": 3900881.0,
|
|
"step": 477
|
|
},
|
|
{
|
|
"entropy": 0.4371698275208473,
|
|
"epoch": 0.2090759978130126,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 7.492186701364007e-08,
|
|
"loss": 0.4635161757469177,
|
|
"mean_token_accuracy": 0.8470185697078705,
|
|
"num_tokens": 3909073.0,
|
|
"step": 478
|
|
},
|
|
{
|
|
"entropy": 0.5315583720803261,
|
|
"epoch": 0.20951339529797702,
|
|
"grad_norm": 2.546875,
|
|
"learning_rate": 7.072179188262252e-08,
|
|
"loss": 0.567477822303772,
|
|
"mean_token_accuracy": 0.8213587403297424,
|
|
"num_tokens": 3917265.0,
|
|
"step": 479
|
|
},
|
|
{
|
|
"entropy": 0.4574076384305954,
|
|
"epoch": 0.2099507927829415,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 6.664117583883272e-08,
|
|
"loss": 0.4899052381515503,
|
|
"mean_token_accuracy": 0.8353865444660187,
|
|
"num_tokens": 3925304.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 0.5328256785869598,
|
|
"epoch": 0.21038819026790595,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 6.268021954544095e-08,
|
|
"loss": 0.5751959085464478,
|
|
"mean_token_accuracy": 0.8115835636854172,
|
|
"num_tokens": 3933496.0,
|
|
"step": 481
|
|
},
|
|
{
|
|
"entropy": 0.4958766922354698,
|
|
"epoch": 0.2108255877528704,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 5.883911778138324e-08,
|
|
"loss": 0.5280551910400391,
|
|
"mean_token_accuracy": 0.8273460417985916,
|
|
"num_tokens": 3941688.0,
|
|
"step": 482
|
|
},
|
|
{
|
|
"entropy": 0.46615899354219437,
|
|
"epoch": 0.21126298523783488,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 5.5118059431781e-08,
|
|
"loss": 0.4785313904285431,
|
|
"mean_token_accuracy": 0.8441089540719986,
|
|
"num_tokens": 3949774.0,
|
|
"step": 483
|
|
},
|
|
{
|
|
"entropy": 0.5289159864187241,
|
|
"epoch": 0.21170038272279934,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 5.151722747865434e-08,
|
|
"loss": 0.5834380984306335,
|
|
"mean_token_accuracy": 0.8157380223274231,
|
|
"num_tokens": 3957966.0,
|
|
"step": 484
|
|
},
|
|
{
|
|
"entropy": 0.4474088177084923,
|
|
"epoch": 0.2121377802077638,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 4.8036798991923925e-08,
|
|
"loss": 0.49255403876304626,
|
|
"mean_token_accuracy": 0.8377321660518646,
|
|
"num_tokens": 3966158.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 0.4575616270303726,
|
|
"epoch": 0.21257517769272827,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 4.467694512070242e-08,
|
|
"loss": 0.48688727617263794,
|
|
"mean_token_accuracy": 0.8357771188020706,
|
|
"num_tokens": 3974350.0,
|
|
"step": 486
|
|
},
|
|
{
|
|
"entropy": 0.44955013692379,
|
|
"epoch": 0.21301257517769273,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 4.1437831084878974e-08,
|
|
"loss": 0.47580015659332275,
|
|
"mean_token_accuracy": 0.8404203206300735,
|
|
"num_tokens": 3982542.0,
|
|
"step": 487
|
|
},
|
|
{
|
|
"entropy": 0.4375215768814087,
|
|
"epoch": 0.2134499726626572,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 3.831961616699464e-08,
|
|
"loss": 0.45662811398506165,
|
|
"mean_token_accuracy": 0.8451857268810272,
|
|
"num_tokens": 3990734.0,
|
|
"step": 488
|
|
},
|
|
{
|
|
"entropy": 0.4607667848467827,
|
|
"epoch": 0.21388737014762166,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 3.5322453704410286e-08,
|
|
"loss": 0.501000165939331,
|
|
"mean_token_accuracy": 0.8324779868125916,
|
|
"num_tokens": 3998926.0,
|
|
"step": 489
|
|
},
|
|
{
|
|
"entropy": 0.4871944785118103,
|
|
"epoch": 0.21432476763258612,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 3.2446491081764566e-08,
|
|
"loss": 0.5080267786979675,
|
|
"mean_token_accuracy": 0.8289345055818558,
|
|
"num_tokens": 4007118.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 0.48445210605859756,
|
|
"epoch": 0.21476216511755059,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 2.969186972372806e-08,
|
|
"loss": 0.5219994187355042,
|
|
"mean_token_accuracy": 0.8257575780153275,
|
|
"num_tokens": 4015310.0,
|
|
"step": 491
|
|
},
|
|
{
|
|
"entropy": 0.5076911747455597,
|
|
"epoch": 0.21519956260251505,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 2.7058725088047466e-08,
|
|
"loss": 0.5234426856040955,
|
|
"mean_token_accuracy": 0.8255131840705872,
|
|
"num_tokens": 4023502.0,
|
|
"step": 492
|
|
},
|
|
{
|
|
"entropy": 0.48257502168416977,
|
|
"epoch": 0.21563696008747948,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 2.454718665888589e-08,
|
|
"loss": 0.5169442296028137,
|
|
"mean_token_accuracy": 0.8269111812114716,
|
|
"num_tokens": 4031419.0,
|
|
"step": 493
|
|
},
|
|
{
|
|
"entropy": 0.48920300602912903,
|
|
"epoch": 0.21607435757244395,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 2.215737794045436e-08,
|
|
"loss": 0.5300370454788208,
|
|
"mean_token_accuracy": 0.8225806355476379,
|
|
"num_tokens": 4039611.0,
|
|
"step": 494
|
|
},
|
|
{
|
|
"entropy": 0.5572062358260155,
|
|
"epoch": 0.2165117550574084,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 1.9889416450938335e-08,
|
|
"loss": 0.5881485939025879,
|
|
"mean_token_accuracy": 0.8054741024971008,
|
|
"num_tokens": 4047803.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 0.42473385483026505,
|
|
"epoch": 0.21694915254237288,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 1.7743413716720394e-08,
|
|
"loss": 0.44505032896995544,
|
|
"mean_token_accuracy": 0.8527614921331406,
|
|
"num_tokens": 4055995.0,
|
|
"step": 496
|
|
},
|
|
{
|
|
"entropy": 0.5073227211833,
|
|
"epoch": 0.21738655002733734,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 1.571947526689349e-08,
|
|
"loss": 0.5384668707847595,
|
|
"mean_token_accuracy": 0.8238025307655334,
|
|
"num_tokens": 4064187.0,
|
|
"step": 497
|
|
},
|
|
{
|
|
"entropy": 0.5434798374772072,
|
|
"epoch": 0.2178239475123018,
|
|
"grad_norm": 2.796875,
|
|
"learning_rate": 1.3817700628074582e-08,
|
|
"loss": 0.5646654367446899,
|
|
"mean_token_accuracy": 0.8227028399705887,
|
|
"num_tokens": 4072379.0,
|
|
"step": 498
|
|
},
|
|
{
|
|
"entropy": 0.40285155177116394,
|
|
"epoch": 0.21826134499726627,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 1.2038183319507957e-08,
|
|
"loss": 0.43041908740997314,
|
|
"mean_token_accuracy": 0.85713692009449,
|
|
"num_tokens": 4080489.0,
|
|
"step": 499
|
|
},
|
|
{
|
|
"entropy": 0.5109951198101044,
|
|
"epoch": 0.21869874248223073,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 1.0381010848466988e-08,
|
|
"loss": 0.5344597101211548,
|
|
"mean_token_accuracy": 0.823924720287323,
|
|
"num_tokens": 4088681.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 0.5647617131471634,
|
|
"epoch": 0.2191361399671952,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 8.84626470595229e-09,
|
|
"loss": 0.5811930894851685,
|
|
"mean_token_accuracy": 0.8090175986289978,
|
|
"num_tokens": 4096873.0,
|
|
"step": 501
|
|
},
|
|
{
|
|
"entropy": 0.5019998624920845,
|
|
"epoch": 0.21957353745215966,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 7.434020362682703e-09,
|
|
"loss": 0.5267874598503113,
|
|
"mean_token_accuracy": 0.8250655829906464,
|
|
"num_tokens": 4104919.0,
|
|
"step": 502
|
|
},
|
|
{
|
|
"entropy": 0.5021177157759666,
|
|
"epoch": 0.22001093493712412,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 6.144347265384931e-09,
|
|
"loss": 0.5334129929542542,
|
|
"mean_token_accuracy": 0.8255132138729095,
|
|
"num_tokens": 4113111.0,
|
|
"step": 503
|
|
},
|
|
{
|
|
"entropy": 0.4931086599826813,
|
|
"epoch": 0.22044833242208858,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 4.977308833379324e-09,
|
|
"loss": 0.5266363024711609,
|
|
"mean_token_accuracy": 0.828323557972908,
|
|
"num_tokens": 4121303.0,
|
|
"step": 504
|
|
},
|
|
{
|
|
"entropy": 0.4733823910355568,
|
|
"epoch": 0.22088572990705305,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 3.932962455458489e-09,
|
|
"loss": 0.5121600031852722,
|
|
"mean_token_accuracy": 0.8324780017137527,
|
|
"num_tokens": 4129495.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 0.45189718902111053,
|
|
"epoch": 0.22132312739201748,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 3.011359487068988e-09,
|
|
"loss": 0.4659193158149719,
|
|
"mean_token_accuracy": 0.8465298265218735,
|
|
"num_tokens": 4137687.0,
|
|
"step": 506
|
|
},
|
|
{
|
|
"entropy": 0.4035600870847702,
|
|
"epoch": 0.22176052487698195,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 2.2125452477828047e-09,
|
|
"loss": 0.41255855560302734,
|
|
"mean_token_accuracy": 0.8647360801696777,
|
|
"num_tokens": 4145879.0,
|
|
"step": 507
|
|
},
|
|
{
|
|
"entropy": 0.5318427979946136,
|
|
"epoch": 0.2221979223619464,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 1.5365590190699608e-09,
|
|
"loss": 0.5627883672714233,
|
|
"mean_token_accuracy": 0.8196481019258499,
|
|
"num_tokens": 4154071.0,
|
|
"step": 508
|
|
},
|
|
{
|
|
"entropy": 0.4271271377801895,
|
|
"epoch": 0.22263531984691087,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 9.834340423678368e-10,
|
|
"loss": 0.4386692941188812,
|
|
"mean_token_accuracy": 0.8549609035253525,
|
|
"num_tokens": 4162263.0,
|
|
"step": 509
|
|
},
|
|
{
|
|
"entropy": 0.5523240566253662,
|
|
"epoch": 0.22307271733187534,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 5.531975174444259e-10,
|
|
"loss": 0.5719492435455322,
|
|
"mean_token_accuracy": 0.8136608004570007,
|
|
"num_tokens": 4170455.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 0.4898983985185623,
|
|
"epoch": 0.2235101148168398,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 2.45870601062459e-10,
|
|
"loss": 0.5295259952545166,
|
|
"mean_token_accuracy": 0.8220918923616409,
|
|
"num_tokens": 4178647.0,
|
|
"step": 511
|
|
},
|
|
{
|
|
"entropy": 0.4591166228055954,
|
|
"epoch": 0.22394751230180426,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 6.14684059385695e-11,
|
|
"loss": 0.49673986434936523,
|
|
"mean_token_accuracy": 0.8360214978456497,
|
|
"num_tokens": 4186839.0,
|
|
"step": 512
|
|
},
|
|
{
|
|
"epoch": 0.22394751230180426,
|
|
"eval_entropy": 0.4859546652017546,
|
|
"eval_loss": 0.5158528089523315,
|
|
"eval_mean_token_accuracy": 0.8297856922977227,
|
|
"eval_num_tokens": 4186839.0,
|
|
"eval_runtime": 32.867,
|
|
"eval_samples_per_second": 29.3,
|
|
"eval_steps_per_second": 3.682,
|
|
"step": 512
|
|
}
|
|
],
|
|
"logging_steps": 1,
|
|
"max_steps": 512,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"save_steps": 128,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": true
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 4.329857209223578e+16,
|
|
"train_batch_size": 2,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|