2c91d89231
Model: fpadovani/swa-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed10 Source: Original Platform
2046 lines
54 KiB
JSON
2046 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.2827254735651682,
|
|
"eval_steps": 1000,
|
|
"global_step": 1000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.838997936248779,
|
|
"epoch": 0.0014136273678258412,
|
|
"grad_norm": 13.1875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 13.9198,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 4340.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.843051767349243,
|
|
"epoch": 0.0028272547356516823,
|
|
"grad_norm": 14.0625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 13.831,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 8692.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.923465204238892,
|
|
"epoch": 0.004240882103477523,
|
|
"grad_norm": 16.75,
|
|
"learning_rate": 7e-06,
|
|
"loss": 13.4913,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 12453.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 5.232606410980225,
|
|
"epoch": 0.005654509471303365,
|
|
"grad_norm": 26.25,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 12.9423,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 16439.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.618780517578125,
|
|
"epoch": 0.007068136839129205,
|
|
"grad_norm": 17.75,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 11.494,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 20515.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.471516036987305,
|
|
"epoch": 0.008481764206955046,
|
|
"grad_norm": 4.34375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.526,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 24336.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.700703144073486,
|
|
"epoch": 0.009895391574780888,
|
|
"grad_norm": 3.8125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.3432,
|
|
"mean_token_accuracy": 0.013547292444854975,
|
|
"num_tokens": 28465.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.689900112152099,
|
|
"epoch": 0.01130901894260673,
|
|
"grad_norm": 3.578125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.9111,
|
|
"mean_token_accuracy": 0.046335773542523384,
|
|
"num_tokens": 32217.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.446650981903076,
|
|
"epoch": 0.01272264631043257,
|
|
"grad_norm": 3.4375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.6264,
|
|
"mean_token_accuracy": 0.040803880989551546,
|
|
"num_tokens": 36369.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.473936176300048,
|
|
"epoch": 0.01413627367825841,
|
|
"grad_norm": 2.703125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.3506,
|
|
"mean_token_accuracy": 0.04644691050052643,
|
|
"num_tokens": 40227.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.520126819610596,
|
|
"epoch": 0.015549901046084252,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.2423,
|
|
"mean_token_accuracy": 0.060035499185323714,
|
|
"num_tokens": 44266.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.462940120697022,
|
|
"epoch": 0.016963528413910092,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.1768,
|
|
"mean_token_accuracy": 0.05174140147864818,
|
|
"num_tokens": 48187.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.422643947601319,
|
|
"epoch": 0.018377155781735936,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.0365,
|
|
"mean_token_accuracy": 0.04872686117887497,
|
|
"num_tokens": 52252.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.438050460815429,
|
|
"epoch": 0.019790783149561775,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 8.9607,
|
|
"mean_token_accuracy": 0.05605476088821888,
|
|
"num_tokens": 56421.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.400554084777832,
|
|
"epoch": 0.021204410517387615,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 8.878,
|
|
"mean_token_accuracy": 0.06232075840234756,
|
|
"num_tokens": 60341.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.250280380249023,
|
|
"epoch": 0.02261803788521346,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 8.7283,
|
|
"mean_token_accuracy": 0.062391096726059914,
|
|
"num_tokens": 64465.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.2134352684021,
|
|
"epoch": 0.0240316652530393,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 8.7126,
|
|
"mean_token_accuracy": 0.06613181754946709,
|
|
"num_tokens": 68514.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.055916213989258,
|
|
"epoch": 0.02544529262086514,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.5801,
|
|
"mean_token_accuracy": 0.06724648177623749,
|
|
"num_tokens": 72661.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 9.991343212127685,
|
|
"epoch": 0.02685891998869098,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.4773,
|
|
"mean_token_accuracy": 0.0715904712677002,
|
|
"num_tokens": 76698.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 9.897418022155762,
|
|
"epoch": 0.02827254735651682,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.3786,
|
|
"mean_token_accuracy": 0.06964126601815224,
|
|
"num_tokens": 80732.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 9.81051197052002,
|
|
"epoch": 0.029686174724342665,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.2664,
|
|
"mean_token_accuracy": 0.07291285954415798,
|
|
"num_tokens": 84797.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 9.750697708129882,
|
|
"epoch": 0.031099802092168505,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.1699,
|
|
"mean_token_accuracy": 0.06931682229042054,
|
|
"num_tokens": 88758.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.548874855041504,
|
|
"epoch": 0.032513429459994345,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.0496,
|
|
"mean_token_accuracy": 0.07683553956449032,
|
|
"num_tokens": 92722.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.418696975708007,
|
|
"epoch": 0.033927056827820185,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 7.8938,
|
|
"mean_token_accuracy": 0.07645067945122719,
|
|
"num_tokens": 96602.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.192409229278564,
|
|
"epoch": 0.035340684195646024,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.831,
|
|
"mean_token_accuracy": 0.07514876201748848,
|
|
"num_tokens": 100598.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.073713493347167,
|
|
"epoch": 0.03675431156347187,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.734,
|
|
"mean_token_accuracy": 0.07320134155452251,
|
|
"num_tokens": 104460.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 8.958717918395996,
|
|
"epoch": 0.03816793893129771,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.6242,
|
|
"mean_token_accuracy": 0.07055199854075908,
|
|
"num_tokens": 108360.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 8.655115032196045,
|
|
"epoch": 0.03958156629912355,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.5081,
|
|
"mean_token_accuracy": 0.07828053757548332,
|
|
"num_tokens": 112176.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.63590440750122,
|
|
"epoch": 0.04099519366694939,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.6179,
|
|
"mean_token_accuracy": 0.06859740577638149,
|
|
"num_tokens": 116192.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.526045417785644,
|
|
"epoch": 0.04240882103477523,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.5008,
|
|
"mean_token_accuracy": 0.06685782484710216,
|
|
"num_tokens": 120221.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.428415966033935,
|
|
"epoch": 0.04382244840260108,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.5019,
|
|
"mean_token_accuracy": 0.06606126651167869,
|
|
"num_tokens": 124364.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.274092197418213,
|
|
"epoch": 0.04523607577042692,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.3469,
|
|
"mean_token_accuracy": 0.08038331940770149,
|
|
"num_tokens": 128266.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.154208850860595,
|
|
"epoch": 0.04664970313825276,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.3261,
|
|
"mean_token_accuracy": 0.07359547801315784,
|
|
"num_tokens": 131985.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.039358854293823,
|
|
"epoch": 0.0480633305060786,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.2318,
|
|
"mean_token_accuracy": 0.08298368901014327,
|
|
"num_tokens": 135892.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.960572910308838,
|
|
"epoch": 0.04947695787390444,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.3743,
|
|
"mean_token_accuracy": 0.07648940533399581,
|
|
"num_tokens": 139937.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.018472337722779,
|
|
"epoch": 0.05089058524173028,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.1981,
|
|
"mean_token_accuracy": 0.08083997145295144,
|
|
"num_tokens": 143659.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.916880559921265,
|
|
"epoch": 0.052304212609556124,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.3147,
|
|
"mean_token_accuracy": 0.08173628598451614,
|
|
"num_tokens": 147394.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.840789651870727,
|
|
"epoch": 0.05371783997738196,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.1724,
|
|
"mean_token_accuracy": 0.08405760079622268,
|
|
"num_tokens": 151082.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.727948999404907,
|
|
"epoch": 0.0551314673452078,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.1786,
|
|
"mean_token_accuracy": 0.08063250184059143,
|
|
"num_tokens": 155099.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.822998285293579,
|
|
"epoch": 0.05654509471303364,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.3548,
|
|
"mean_token_accuracy": 0.07048202715814114,
|
|
"num_tokens": 159297.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.713145303726196,
|
|
"epoch": 0.05795872208085948,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.2237,
|
|
"mean_token_accuracy": 0.07353744581341744,
|
|
"num_tokens": 163337.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.855460119247437,
|
|
"epoch": 0.05937234944868533,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.277,
|
|
"mean_token_accuracy": 0.08301040753722191,
|
|
"num_tokens": 167440.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.675248670578003,
|
|
"epoch": 0.06078597681651117,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.1489,
|
|
"mean_token_accuracy": 0.08562113344669342,
|
|
"num_tokens": 171238.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.707080364227295,
|
|
"epoch": 0.06219960418433701,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.2134,
|
|
"mean_token_accuracy": 0.07360663190484047,
|
|
"num_tokens": 175476.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.5691710948944095,
|
|
"epoch": 0.06361323155216285,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.1273,
|
|
"mean_token_accuracy": 0.08262931331992149,
|
|
"num_tokens": 179398.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.5606465339660645,
|
|
"epoch": 0.06502685891998869,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.1102,
|
|
"mean_token_accuracy": 0.08324644044041633,
|
|
"num_tokens": 183355.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.605392646789551,
|
|
"epoch": 0.06644048628781453,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.12,
|
|
"mean_token_accuracy": 0.07809790521860123,
|
|
"num_tokens": 187278.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.519303512573242,
|
|
"epoch": 0.06785411365564037,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.1402,
|
|
"mean_token_accuracy": 0.0761954978108406,
|
|
"num_tokens": 191136.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.577075338363647,
|
|
"epoch": 0.06926774102346621,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.2122,
|
|
"mean_token_accuracy": 0.08024812079966068,
|
|
"num_tokens": 195096.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.606031370162964,
|
|
"epoch": 0.07068136839129205,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.2001,
|
|
"mean_token_accuracy": 0.07799483351409435,
|
|
"num_tokens": 199483.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.643957281112671,
|
|
"epoch": 0.0720949957591179,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1263,
|
|
"mean_token_accuracy": 0.08172452189028263,
|
|
"num_tokens": 203515.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.497114038467407,
|
|
"epoch": 0.07350862312694374,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.1289,
|
|
"mean_token_accuracy": 0.07867333814501762,
|
|
"num_tokens": 207451.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.548718070983886,
|
|
"epoch": 0.07492225049476958,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.0456,
|
|
"mean_token_accuracy": 0.08255640640854836,
|
|
"num_tokens": 211507.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.439682149887085,
|
|
"epoch": 0.07633587786259542,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.1149,
|
|
"mean_token_accuracy": 0.08738862499594688,
|
|
"num_tokens": 215236.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.497509336471557,
|
|
"epoch": 0.07774950523042126,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.0841,
|
|
"mean_token_accuracy": 0.08626129925251007,
|
|
"num_tokens": 218934.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.530088710784912,
|
|
"epoch": 0.0791631325982471,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.031,
|
|
"mean_token_accuracy": 0.08123192861676216,
|
|
"num_tokens": 223185.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.557523965835571,
|
|
"epoch": 0.08057675996607294,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.1551,
|
|
"mean_token_accuracy": 0.07593904286623002,
|
|
"num_tokens": 227044.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.449309587478638,
|
|
"epoch": 0.08199038733389878,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.0628,
|
|
"mean_token_accuracy": 0.09167415499687195,
|
|
"num_tokens": 231359.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.531680345535278,
|
|
"epoch": 0.08340401470172462,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.0471,
|
|
"mean_token_accuracy": 0.08983734510838985,
|
|
"num_tokens": 235224.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.344621992111206,
|
|
"epoch": 0.08481764206955046,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.905,
|
|
"mean_token_accuracy": 0.09648748189210891,
|
|
"num_tokens": 239119.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.288154172897339,
|
|
"epoch": 0.0862312694373763,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.0089,
|
|
"mean_token_accuracy": 0.0949060007929802,
|
|
"num_tokens": 243275.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.433500814437866,
|
|
"epoch": 0.08764489680520215,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.0206,
|
|
"mean_token_accuracy": 0.09643495455384254,
|
|
"num_tokens": 247415.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.416780138015747,
|
|
"epoch": 0.089058524173028,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.0159,
|
|
"mean_token_accuracy": 0.08751917779445648,
|
|
"num_tokens": 251575.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.372589826583862,
|
|
"epoch": 0.09047215154085383,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.9997,
|
|
"mean_token_accuracy": 0.08504554927349091,
|
|
"num_tokens": 255686.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.3249767303466795,
|
|
"epoch": 0.09188577890867967,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.9852,
|
|
"mean_token_accuracy": 0.09507602602243423,
|
|
"num_tokens": 259850.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.423571062088013,
|
|
"epoch": 0.09329940627650551,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.0103,
|
|
"mean_token_accuracy": 0.09533059746026992,
|
|
"num_tokens": 264043.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.514725065231323,
|
|
"epoch": 0.09471303364433135,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.0725,
|
|
"mean_token_accuracy": 0.08970947004854679,
|
|
"num_tokens": 268296.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.174638795852661,
|
|
"epoch": 0.0961266610121572,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.8961,
|
|
"mean_token_accuracy": 0.09778379574418068,
|
|
"num_tokens": 272062.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.25974817276001,
|
|
"epoch": 0.09754028837998303,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.8514,
|
|
"mean_token_accuracy": 0.10595791339874268,
|
|
"num_tokens": 275936.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.272134685516358,
|
|
"epoch": 0.09895391574780887,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.8802,
|
|
"mean_token_accuracy": 0.10179490596055984,
|
|
"num_tokens": 279526.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.339320135116577,
|
|
"epoch": 0.10036754311563471,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.9552,
|
|
"mean_token_accuracy": 0.0998102031648159,
|
|
"num_tokens": 283577.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.324524545669556,
|
|
"epoch": 0.10178117048346055,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.7984,
|
|
"mean_token_accuracy": 0.101173298060894,
|
|
"num_tokens": 287377.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.140333604812622,
|
|
"epoch": 0.10319479785128641,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.973,
|
|
"mean_token_accuracy": 0.09467578828334808,
|
|
"num_tokens": 291582.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.392356729507446,
|
|
"epoch": 0.10460842521911225,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.8489,
|
|
"mean_token_accuracy": 0.0993436373770237,
|
|
"num_tokens": 295608.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.154901552200317,
|
|
"epoch": 0.10602205258693809,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.872,
|
|
"mean_token_accuracy": 0.1050638273358345,
|
|
"num_tokens": 299660.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.05282769203186,
|
|
"epoch": 0.10743567995476393,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.7846,
|
|
"mean_token_accuracy": 0.11004708558321,
|
|
"num_tokens": 303412.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.143272113800049,
|
|
"epoch": 0.10884930732258977,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.8498,
|
|
"mean_token_accuracy": 0.10617732927203179,
|
|
"num_tokens": 307352.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.2218762874603275,
|
|
"epoch": 0.1102629346904156,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.9129,
|
|
"mean_token_accuracy": 0.10999525636434555,
|
|
"num_tokens": 311418.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.1364624977111815,
|
|
"epoch": 0.11167656205824145,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.7597,
|
|
"mean_token_accuracy": 0.11771088987588882,
|
|
"num_tokens": 315150.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.207963132858277,
|
|
"epoch": 0.11309018942606729,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.9407,
|
|
"mean_token_accuracy": 0.10095959007740021,
|
|
"num_tokens": 318898.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.126479911804199,
|
|
"epoch": 0.11450381679389313,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.7726,
|
|
"mean_token_accuracy": 0.11357507780194283,
|
|
"num_tokens": 322495.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.224441242218018,
|
|
"epoch": 0.11591744416171897,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.8993,
|
|
"mean_token_accuracy": 0.11550437733530998,
|
|
"num_tokens": 326473.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.140273571014404,
|
|
"epoch": 0.1173310715295448,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.8522,
|
|
"mean_token_accuracy": 0.10760911256074905,
|
|
"num_tokens": 330393.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.134380578994751,
|
|
"epoch": 0.11874469889737066,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.7692,
|
|
"mean_token_accuracy": 0.11172740682959556,
|
|
"num_tokens": 334449.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.2949425220489506,
|
|
"epoch": 0.1201583262651965,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.9219,
|
|
"mean_token_accuracy": 0.10026353597640991,
|
|
"num_tokens": 338740.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.066179800033569,
|
|
"epoch": 0.12157195363302234,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.7336,
|
|
"mean_token_accuracy": 0.11876392140984535,
|
|
"num_tokens": 342429.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.117091083526612,
|
|
"epoch": 0.12298558100084818,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.7741,
|
|
"mean_token_accuracy": 0.12331236079335213,
|
|
"num_tokens": 346218.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.182254219055176,
|
|
"epoch": 0.12439920836867402,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.7754,
|
|
"mean_token_accuracy": 0.11808006688952447,
|
|
"num_tokens": 350310.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.058240985870361,
|
|
"epoch": 0.12581283573649985,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.7434,
|
|
"mean_token_accuracy": 0.12305478826165199,
|
|
"num_tokens": 353952.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.198861885070801,
|
|
"epoch": 0.1272264631043257,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.9148,
|
|
"mean_token_accuracy": 0.11514305397868156,
|
|
"num_tokens": 357667.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.127170419692993,
|
|
"epoch": 0.12864009047215155,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.7165,
|
|
"mean_token_accuracy": 0.12080714330077172,
|
|
"num_tokens": 361309.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.963030576705933,
|
|
"epoch": 0.13005371783997738,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.7651,
|
|
"mean_token_accuracy": 0.13394678458571435,
|
|
"num_tokens": 365099.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.293617916107178,
|
|
"epoch": 0.13146734520780323,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.8263,
|
|
"mean_token_accuracy": 0.11883056983351707,
|
|
"num_tokens": 369141.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.0113513469696045,
|
|
"epoch": 0.13288097257562906,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.7673,
|
|
"mean_token_accuracy": 0.11753190383315086,
|
|
"num_tokens": 373128.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.056188488006592,
|
|
"epoch": 0.1342945999434549,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.7551,
|
|
"mean_token_accuracy": 0.12295122221112251,
|
|
"num_tokens": 377103.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.010499048233032,
|
|
"epoch": 0.13570822731128074,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.8054,
|
|
"mean_token_accuracy": 0.11294787377119064,
|
|
"num_tokens": 381573.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.241208553314209,
|
|
"epoch": 0.1371218546791066,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.8248,
|
|
"mean_token_accuracy": 0.12627912536263466,
|
|
"num_tokens": 385759.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.94762954711914,
|
|
"epoch": 0.13853548204693242,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.7048,
|
|
"mean_token_accuracy": 0.12667714580893516,
|
|
"num_tokens": 389700.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.064765691757202,
|
|
"epoch": 0.13994910941475827,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.6205,
|
|
"mean_token_accuracy": 0.12440077215433121,
|
|
"num_tokens": 393279.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.031151533126831,
|
|
"epoch": 0.1413627367825841,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.725,
|
|
"mean_token_accuracy": 0.12710191905498505,
|
|
"num_tokens": 397379.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.966974687576294,
|
|
"epoch": 0.14277636415040995,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.7762,
|
|
"mean_token_accuracy": 0.12484551221132278,
|
|
"num_tokens": 401349.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.063830709457397,
|
|
"epoch": 0.1441899915182358,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.6911,
|
|
"mean_token_accuracy": 0.12564558237791063,
|
|
"num_tokens": 405361.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.972732162475586,
|
|
"epoch": 0.14560361888606163,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.726,
|
|
"mean_token_accuracy": 0.12524131089448928,
|
|
"num_tokens": 409614.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.847900104522705,
|
|
"epoch": 0.14701724625388748,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.6623,
|
|
"mean_token_accuracy": 0.13759828507900237,
|
|
"num_tokens": 413529.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.994072675704956,
|
|
"epoch": 0.1484308736217133,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.6178,
|
|
"mean_token_accuracy": 0.12413196116685868,
|
|
"num_tokens": 417365.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.847810077667236,
|
|
"epoch": 0.14984450098953916,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.6269,
|
|
"mean_token_accuracy": 0.1389990709722042,
|
|
"num_tokens": 421232.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.125303459167481,
|
|
"epoch": 0.151258128357365,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.7183,
|
|
"mean_token_accuracy": 0.12456061840057372,
|
|
"num_tokens": 425503.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.852126216888427,
|
|
"epoch": 0.15267175572519084,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.7542,
|
|
"mean_token_accuracy": 0.12805211022496224,
|
|
"num_tokens": 429770.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.0629345893859865,
|
|
"epoch": 0.15408538309301667,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.6387,
|
|
"mean_token_accuracy": 0.13042709827423096,
|
|
"num_tokens": 433657.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.941704559326172,
|
|
"epoch": 0.15549901046084252,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.7511,
|
|
"mean_token_accuracy": 0.1230709470808506,
|
|
"num_tokens": 437732.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.886470556259155,
|
|
"epoch": 0.15691263782866835,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.4963,
|
|
"mean_token_accuracy": 0.13797944858670236,
|
|
"num_tokens": 441866.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.786446952819825,
|
|
"epoch": 0.1583262651964942,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.5783,
|
|
"mean_token_accuracy": 0.1337943233549595,
|
|
"num_tokens": 445437.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.786550521850586,
|
|
"epoch": 0.15973989256432006,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.564,
|
|
"mean_token_accuracy": 0.13779796212911605,
|
|
"num_tokens": 449602.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.764790821075439,
|
|
"epoch": 0.16115351993214588,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.4694,
|
|
"mean_token_accuracy": 0.13991471529006957,
|
|
"num_tokens": 453140.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.97475233078003,
|
|
"epoch": 0.16256714729997174,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.7096,
|
|
"mean_token_accuracy": 0.12624734565615653,
|
|
"num_tokens": 457210.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.009247064590454,
|
|
"epoch": 0.16398077466779756,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.6826,
|
|
"mean_token_accuracy": 0.12368027344346047,
|
|
"num_tokens": 461534.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.9301024913787845,
|
|
"epoch": 0.16539440203562342,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.4802,
|
|
"mean_token_accuracy": 0.13896113261580467,
|
|
"num_tokens": 465055.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.829527807235718,
|
|
"epoch": 0.16680802940344924,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.5188,
|
|
"mean_token_accuracy": 0.13359609320759774,
|
|
"num_tokens": 469133.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.923969316482544,
|
|
"epoch": 0.1682216567712751,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.6646,
|
|
"mean_token_accuracy": 0.1328178010880947,
|
|
"num_tokens": 473078.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.906017303466797,
|
|
"epoch": 0.16963528413910092,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.5999,
|
|
"mean_token_accuracy": 0.13147619739174843,
|
|
"num_tokens": 476988.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.845853567123413,
|
|
"epoch": 0.17104891150692678,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.4638,
|
|
"mean_token_accuracy": 0.13877335488796233,
|
|
"num_tokens": 481170.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.808208322525024,
|
|
"epoch": 0.1724625388747526,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.4607,
|
|
"mean_token_accuracy": 0.12855389714241028,
|
|
"num_tokens": 484853.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.694374704360962,
|
|
"epoch": 0.17387616624257846,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.5042,
|
|
"mean_token_accuracy": 0.14368257224559783,
|
|
"num_tokens": 488550.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.714710140228272,
|
|
"epoch": 0.1752897936104043,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.5511,
|
|
"mean_token_accuracy": 0.1393193356692791,
|
|
"num_tokens": 492621.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.823784923553466,
|
|
"epoch": 0.17670342097823014,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.4969,
|
|
"mean_token_accuracy": 0.12763671800494195,
|
|
"num_tokens": 496627.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.75352110862732,
|
|
"epoch": 0.178117048346056,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.5534,
|
|
"mean_token_accuracy": 0.13516684994101524,
|
|
"num_tokens": 500794.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.771178388595581,
|
|
"epoch": 0.17953067571388182,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.5715,
|
|
"mean_token_accuracy": 0.1311728797852993,
|
|
"num_tokens": 504672.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.785078620910644,
|
|
"epoch": 0.18094430308170767,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.4995,
|
|
"mean_token_accuracy": 0.14259493052959443,
|
|
"num_tokens": 508834.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.843728113174438,
|
|
"epoch": 0.1823579304495335,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.4686,
|
|
"mean_token_accuracy": 0.13577061742544175,
|
|
"num_tokens": 512722.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.774205303192138,
|
|
"epoch": 0.18377155781735935,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.5824,
|
|
"mean_token_accuracy": 0.13288535103201865,
|
|
"num_tokens": 517270.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.76011438369751,
|
|
"epoch": 0.18518518518518517,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.5048,
|
|
"mean_token_accuracy": 0.14277215600013732,
|
|
"num_tokens": 521438.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.762216997146607,
|
|
"epoch": 0.18659881255301103,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.4822,
|
|
"mean_token_accuracy": 0.1396136611700058,
|
|
"num_tokens": 525890.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.725051021575927,
|
|
"epoch": 0.18801243992083685,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.4927,
|
|
"mean_token_accuracy": 0.14150392562150954,
|
|
"num_tokens": 529775.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.711924600601196,
|
|
"epoch": 0.1894260672886627,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.4771,
|
|
"mean_token_accuracy": 0.14005500972270965,
|
|
"num_tokens": 533889.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.584859466552734,
|
|
"epoch": 0.19083969465648856,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.402,
|
|
"mean_token_accuracy": 0.1481585830450058,
|
|
"num_tokens": 537701.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.701951599121093,
|
|
"epoch": 0.1922533220243144,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.4879,
|
|
"mean_token_accuracy": 0.14182492196559907,
|
|
"num_tokens": 542160.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.702091026306152,
|
|
"epoch": 0.19366694939214024,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.4397,
|
|
"mean_token_accuracy": 0.14451163560152053,
|
|
"num_tokens": 546317.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.670078468322754,
|
|
"epoch": 0.19508057675996607,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.2943,
|
|
"mean_token_accuracy": 0.15481528118252755,
|
|
"num_tokens": 550304.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.529588890075684,
|
|
"epoch": 0.19649420412779192,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.3855,
|
|
"mean_token_accuracy": 0.1449325367808342,
|
|
"num_tokens": 554276.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.679137516021728,
|
|
"epoch": 0.19790783149561775,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.4109,
|
|
"mean_token_accuracy": 0.144695745408535,
|
|
"num_tokens": 558044.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.46890025138855,
|
|
"epoch": 0.1993214588634436,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.398,
|
|
"mean_token_accuracy": 0.15260268077254296,
|
|
"num_tokens": 561667.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.7185471534729,
|
|
"epoch": 0.20073508623126943,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.421,
|
|
"mean_token_accuracy": 0.1382300853729248,
|
|
"num_tokens": 565738.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.691477727890015,
|
|
"epoch": 0.20214871359909528,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.4849,
|
|
"mean_token_accuracy": 0.1443715550005436,
|
|
"num_tokens": 569629.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.537031507492065,
|
|
"epoch": 0.2035623409669211,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.3245,
|
|
"mean_token_accuracy": 0.14985155463218688,
|
|
"num_tokens": 573272.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.596857976913452,
|
|
"epoch": 0.20497596833474696,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.3822,
|
|
"mean_token_accuracy": 0.14262079298496247,
|
|
"num_tokens": 577224.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.525388908386231,
|
|
"epoch": 0.20638959570257281,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.3224,
|
|
"mean_token_accuracy": 0.15136756971478463,
|
|
"num_tokens": 580919.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.716662502288818,
|
|
"epoch": 0.20780322307039864,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.445,
|
|
"mean_token_accuracy": 0.1410931281745434,
|
|
"num_tokens": 584985.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.596371221542358,
|
|
"epoch": 0.2092168504382245,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.3992,
|
|
"mean_token_accuracy": 0.15255053117871284,
|
|
"num_tokens": 588979.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.564611911773682,
|
|
"epoch": 0.21063047780605032,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.3667,
|
|
"mean_token_accuracy": 0.1505613371729851,
|
|
"num_tokens": 592930.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.536220741271973,
|
|
"epoch": 0.21204410517387617,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.4103,
|
|
"mean_token_accuracy": 0.1515594244003296,
|
|
"num_tokens": 596832.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.828583717346191,
|
|
"epoch": 0.213457732541702,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.446,
|
|
"mean_token_accuracy": 0.14928378835320472,
|
|
"num_tokens": 600934.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.329220914840699,
|
|
"epoch": 0.21487135990952785,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.3199,
|
|
"mean_token_accuracy": 0.1495775043964386,
|
|
"num_tokens": 604868.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.744077539443969,
|
|
"epoch": 0.21628498727735368,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.4376,
|
|
"mean_token_accuracy": 0.13565291613340377,
|
|
"num_tokens": 608747.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.577572679519653,
|
|
"epoch": 0.21769861464517953,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.4215,
|
|
"mean_token_accuracy": 0.1479615420103073,
|
|
"num_tokens": 612627.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.517498445510864,
|
|
"epoch": 0.21911224201300536,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.3368,
|
|
"mean_token_accuracy": 0.1571616619825363,
|
|
"num_tokens": 616455.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.5876189231872555,
|
|
"epoch": 0.2205258693808312,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.4168,
|
|
"mean_token_accuracy": 0.1469177559018135,
|
|
"num_tokens": 620617.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.415558195114135,
|
|
"epoch": 0.22193949674865707,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.3376,
|
|
"mean_token_accuracy": 0.1490316942334175,
|
|
"num_tokens": 624863.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.5751752853393555,
|
|
"epoch": 0.2233531241164829,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.3625,
|
|
"mean_token_accuracy": 0.15206135734915732,
|
|
"num_tokens": 629073.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.595431232452393,
|
|
"epoch": 0.22476675148430875,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.3723,
|
|
"mean_token_accuracy": 0.15247093737125397,
|
|
"num_tokens": 633550.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.571492719650268,
|
|
"epoch": 0.22618037885213457,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.3611,
|
|
"mean_token_accuracy": 0.15650528520345688,
|
|
"num_tokens": 637459.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.416607141494751,
|
|
"epoch": 0.22759400621996043,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.3272,
|
|
"mean_token_accuracy": 0.14604957774281502,
|
|
"num_tokens": 641587.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.561601161956787,
|
|
"epoch": 0.22900763358778625,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.5026,
|
|
"mean_token_accuracy": 0.14142623618245126,
|
|
"num_tokens": 645726.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.5540244579315186,
|
|
"epoch": 0.2304212609556121,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.2808,
|
|
"mean_token_accuracy": 0.15965329706668854,
|
|
"num_tokens": 649688.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.466880655288696,
|
|
"epoch": 0.23183488832343793,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.3081,
|
|
"mean_token_accuracy": 0.14743991270661355,
|
|
"num_tokens": 653745.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.281443119049072,
|
|
"epoch": 0.23324851569126379,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.2163,
|
|
"mean_token_accuracy": 0.1635961800813675,
|
|
"num_tokens": 657650.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.540815114974976,
|
|
"epoch": 0.2346621430590896,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.3282,
|
|
"mean_token_accuracy": 0.15691596940159797,
|
|
"num_tokens": 661614.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.437383794784546,
|
|
"epoch": 0.23607577042691547,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.2877,
|
|
"mean_token_accuracy": 0.15950870811939238,
|
|
"num_tokens": 665484.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.302047681808472,
|
|
"epoch": 0.23748939779474132,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.3023,
|
|
"mean_token_accuracy": 0.1515549197793007,
|
|
"num_tokens": 669718.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.494394445419312,
|
|
"epoch": 0.23890302516256715,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.2621,
|
|
"mean_token_accuracy": 0.1587437778711319,
|
|
"num_tokens": 673913.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.465457439422607,
|
|
"epoch": 0.240316652530393,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.3427,
|
|
"mean_token_accuracy": 0.14686428755521774,
|
|
"num_tokens": 677911.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.367018795013427,
|
|
"epoch": 0.24173027989821882,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.1437,
|
|
"mean_token_accuracy": 0.16687410622835158,
|
|
"num_tokens": 681629.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.405222368240357,
|
|
"epoch": 0.24314390726604468,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.1789,
|
|
"mean_token_accuracy": 0.15915078967809676,
|
|
"num_tokens": 685675.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.3251574516296385,
|
|
"epoch": 0.2445575346338705,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.1061,
|
|
"mean_token_accuracy": 0.16585489213466645,
|
|
"num_tokens": 689630.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.331725025177002,
|
|
"epoch": 0.24597116200169636,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.2086,
|
|
"mean_token_accuracy": 0.15823635309934617,
|
|
"num_tokens": 693580.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.292018890380859,
|
|
"epoch": 0.24738478936952218,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.2824,
|
|
"mean_token_accuracy": 0.15428755432367325,
|
|
"num_tokens": 697531.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.245914125442505,
|
|
"epoch": 0.24879841673734804,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.2452,
|
|
"mean_token_accuracy": 0.16177606135606765,
|
|
"num_tokens": 701421.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.380286502838135,
|
|
"epoch": 0.2502120441051739,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.2034,
|
|
"mean_token_accuracy": 0.158734093606472,
|
|
"num_tokens": 705557.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.279850482940674,
|
|
"epoch": 0.2516256714729997,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.1344,
|
|
"mean_token_accuracy": 0.1670902654528618,
|
|
"num_tokens": 709173.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.265487957000732,
|
|
"epoch": 0.25303929884082554,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.0852,
|
|
"mean_token_accuracy": 0.1611698791384697,
|
|
"num_tokens": 713144.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.158499622344971,
|
|
"epoch": 0.2544529262086514,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.3942,
|
|
"mean_token_accuracy": 0.15290624499320984,
|
|
"num_tokens": 717187.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.500699520111084,
|
|
"epoch": 0.25586655357647725,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.2656,
|
|
"mean_token_accuracy": 0.14998120591044425,
|
|
"num_tokens": 721311.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.283317804336548,
|
|
"epoch": 0.2572801809443031,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.1865,
|
|
"mean_token_accuracy": 0.1690343588590622,
|
|
"num_tokens": 725193.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.32297830581665,
|
|
"epoch": 0.2586938083121289,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.1775,
|
|
"mean_token_accuracy": 0.1630359873175621,
|
|
"num_tokens": 729208.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.252132749557495,
|
|
"epoch": 0.26010743567995476,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.2431,
|
|
"mean_token_accuracy": 0.162384469807148,
|
|
"num_tokens": 733364.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.240159702301026,
|
|
"epoch": 0.2615210630477806,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.1907,
|
|
"mean_token_accuracy": 0.16134183555841447,
|
|
"num_tokens": 737549.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.357678174972534,
|
|
"epoch": 0.26293469041560646,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.2012,
|
|
"mean_token_accuracy": 0.16150422096252443,
|
|
"num_tokens": 742009.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.26593918800354,
|
|
"epoch": 0.26434831778343226,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.3095,
|
|
"mean_token_accuracy": 0.1559640161693096,
|
|
"num_tokens": 746258.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.422447776794433,
|
|
"epoch": 0.2657619451512581,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.2134,
|
|
"mean_token_accuracy": 0.16972213834524155,
|
|
"num_tokens": 750027.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.179395294189453,
|
|
"epoch": 0.26717557251908397,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.1788,
|
|
"mean_token_accuracy": 0.15804249197244644,
|
|
"num_tokens": 754030.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.166366243362427,
|
|
"epoch": 0.2685891998869098,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.0744,
|
|
"mean_token_accuracy": 0.1662062868475914,
|
|
"num_tokens": 757663.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.173876714706421,
|
|
"epoch": 0.2700028272547357,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.1816,
|
|
"mean_token_accuracy": 0.1622084781527519,
|
|
"num_tokens": 761806.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.358310317993164,
|
|
"epoch": 0.2714164546225615,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.1687,
|
|
"mean_token_accuracy": 0.1678529866039753,
|
|
"num_tokens": 765900.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.13313364982605,
|
|
"epoch": 0.27283008199038733,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.1376,
|
|
"mean_token_accuracy": 0.16201107054948807,
|
|
"num_tokens": 769842.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.349887704849243,
|
|
"epoch": 0.2742437093582132,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.2531,
|
|
"mean_token_accuracy": 0.16530761122703552,
|
|
"num_tokens": 773717.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.021614742279053,
|
|
"epoch": 0.27565733672603904,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 5.9991,
|
|
"mean_token_accuracy": 0.17551461309194566,
|
|
"num_tokens": 777720.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.14023642539978,
|
|
"epoch": 0.27707096409386484,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.027,
|
|
"mean_token_accuracy": 0.16881906613707542,
|
|
"num_tokens": 781482.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.2495396614074705,
|
|
"epoch": 0.2784845914616907,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.1729,
|
|
"mean_token_accuracy": 0.15896565914154054,
|
|
"num_tokens": 785390.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 5.967093658447266,
|
|
"epoch": 0.27989821882951654,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.0915,
|
|
"mean_token_accuracy": 0.1750103384256363,
|
|
"num_tokens": 789464.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.431793308258056,
|
|
"epoch": 0.2813118461973424,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.179,
|
|
"mean_token_accuracy": 0.15985391139984131,
|
|
"num_tokens": 793229.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.186357164382935,
|
|
"epoch": 0.2827254735651682,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.0844,
|
|
"mean_token_accuracy": 0.1675830215215683,
|
|
"num_tokens": 797490.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.2827254735651682,
|
|
"eval_entropy": 6.063302199271112,
|
|
"eval_loss": 6.372450828552246,
|
|
"eval_mean_token_accuracy": 0.16418416006589595,
|
|
"eval_num_tokens": 797490.0,
|
|
"eval_runtime": 6.7832,
|
|
"eval_samples_per_second": 1663.516,
|
|
"eval_steps_per_second": 208.013,
|
|
"step": 1000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 35360,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 1000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 977117884416000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|