10145 lines
276 KiB
JSON
10145 lines
276 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 4.54157201272149,
|
|
"eval_steps": 500,
|
|
"global_step": 5000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742555522918702,
|
|
"epoch": 0.004543389368468878,
|
|
"grad_norm": 5.0625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7779,
|
|
"mean_token_accuracy": 8.431703317910433e-05,
|
|
"num_tokens": 10156.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742563724517822,
|
|
"epoch": 0.009086778736937756,
|
|
"grad_norm": 5.25,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7654,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 20669.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.74255313873291,
|
|
"epoch": 0.013630168105406633,
|
|
"grad_norm": 4.9375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.7397,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 30952.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742581939697265,
|
|
"epoch": 0.01817355747387551,
|
|
"grad_norm": 5.40625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.6785,
|
|
"mean_token_accuracy": 0.0005094278021715581,
|
|
"num_tokens": 41429.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742574501037598,
|
|
"epoch": 0.02271694684234439,
|
|
"grad_norm": 4.90625,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.585,
|
|
"mean_token_accuracy": 0.004951991699635982,
|
|
"num_tokens": 52166.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.742272472381591,
|
|
"epoch": 0.027260336210813266,
|
|
"grad_norm": 4.0,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.439,
|
|
"mean_token_accuracy": 0.027590089477598667,
|
|
"num_tokens": 61670.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.741254997253417,
|
|
"epoch": 0.03180372557928214,
|
|
"grad_norm": 3.203125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.2851,
|
|
"mean_token_accuracy": 0.040197663381695746,
|
|
"num_tokens": 71702.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.738425350189209,
|
|
"epoch": 0.03634711494775102,
|
|
"grad_norm": 2.734375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.129,
|
|
"mean_token_accuracy": 0.04816498160362244,
|
|
"num_tokens": 82840.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.734417152404784,
|
|
"epoch": 0.0408905043162199,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.0098,
|
|
"mean_token_accuracy": 0.04295867718756199,
|
|
"num_tokens": 92290.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.731618976593017,
|
|
"epoch": 0.04543389368468878,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.9604,
|
|
"mean_token_accuracy": 0.04617781527340412,
|
|
"num_tokens": 101842.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.729147243499757,
|
|
"epoch": 0.049977283053157656,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.9546,
|
|
"mean_token_accuracy": 0.04403300806879997,
|
|
"num_tokens": 114030.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.727308368682861,
|
|
"epoch": 0.05452067242162653,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.8581,
|
|
"mean_token_accuracy": 0.04366073869168759,
|
|
"num_tokens": 124349.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.726133060455322,
|
|
"epoch": 0.059064061790095414,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.7412,
|
|
"mean_token_accuracy": 0.044536291435360906,
|
|
"num_tokens": 133534.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.724782562255859,
|
|
"epoch": 0.06360745115856428,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.6971,
|
|
"mean_token_accuracy": 0.041312586329877374,
|
|
"num_tokens": 144383.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.723019886016846,
|
|
"epoch": 0.06815084052703317,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.6751,
|
|
"mean_token_accuracy": 0.04508417844772339,
|
|
"num_tokens": 155467.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.717958164215087,
|
|
"epoch": 0.07269422989550205,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.5919,
|
|
"mean_token_accuracy": 0.045108040794730186,
|
|
"num_tokens": 165903.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.709596824645995,
|
|
"epoch": 0.07723761926397092,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.511,
|
|
"mean_token_accuracy": 0.04253413174301386,
|
|
"num_tokens": 175812.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.702213764190674,
|
|
"epoch": 0.0817810086324398,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.4176,
|
|
"mean_token_accuracy": 0.04180197641253471,
|
|
"num_tokens": 186357.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.6961669921875,
|
|
"epoch": 0.08632439800090867,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.3391,
|
|
"mean_token_accuracy": 0.04544849582016468,
|
|
"num_tokens": 196124.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.684025192260743,
|
|
"epoch": 0.09086778736937756,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.2392,
|
|
"mean_token_accuracy": 0.05092858672142029,
|
|
"num_tokens": 205655.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.664148139953614,
|
|
"epoch": 0.09541117673784644,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.1639,
|
|
"mean_token_accuracy": 0.05325500890612602,
|
|
"num_tokens": 215870.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.64291820526123,
|
|
"epoch": 0.09995456610631531,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.0589,
|
|
"mean_token_accuracy": 0.056853560358285905,
|
|
"num_tokens": 225027.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.60800380706787,
|
|
"epoch": 0.10449795547478419,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.9662,
|
|
"mean_token_accuracy": 0.05881645642220974,
|
|
"num_tokens": 236352.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.564264106750489,
|
|
"epoch": 0.10904134484325306,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.8255,
|
|
"mean_token_accuracy": 0.06017065867781639,
|
|
"num_tokens": 246713.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.5113018989563,
|
|
"epoch": 0.11358473421172194,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.666,
|
|
"mean_token_accuracy": 0.06393568031489849,
|
|
"num_tokens": 256950.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.445667171478272,
|
|
"epoch": 0.11812812358019083,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.6162,
|
|
"mean_token_accuracy": 0.06482643783092498,
|
|
"num_tokens": 267345.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.374823379516602,
|
|
"epoch": 0.1226715129486597,
|
|
"grad_norm": 1.9765625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.4096,
|
|
"mean_token_accuracy": 0.07293099723756313,
|
|
"num_tokens": 276782.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.27814416885376,
|
|
"epoch": 0.12721490231712856,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.2813,
|
|
"mean_token_accuracy": 0.073567845672369,
|
|
"num_tokens": 287024.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.162126541137695,
|
|
"epoch": 0.13175829168559747,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.255,
|
|
"mean_token_accuracy": 0.07345798872411251,
|
|
"num_tokens": 296597.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.039239120483398,
|
|
"epoch": 0.13630168105406634,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.0456,
|
|
"mean_token_accuracy": 0.07541064321994781,
|
|
"num_tokens": 307709.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.879199600219726,
|
|
"epoch": 0.14084507042253522,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.0019,
|
|
"mean_token_accuracy": 0.07952367104589939,
|
|
"num_tokens": 317976.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.700241374969483,
|
|
"epoch": 0.1453884597910041,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.8309,
|
|
"mean_token_accuracy": 0.07648782916367054,
|
|
"num_tokens": 327319.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.498950004577637,
|
|
"epoch": 0.14993184915947297,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.7494,
|
|
"mean_token_accuracy": 0.07844193913042545,
|
|
"num_tokens": 337337.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.23708553314209,
|
|
"epoch": 0.15447523852794184,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.7247,
|
|
"mean_token_accuracy": 0.07220172360539437,
|
|
"num_tokens": 347779.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.071245193481445,
|
|
"epoch": 0.15901862789641072,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.6848,
|
|
"mean_token_accuracy": 0.07456069067120552,
|
|
"num_tokens": 358088.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.767854022979737,
|
|
"epoch": 0.1635620172648796,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.5652,
|
|
"mean_token_accuracy": 0.07797556966543198,
|
|
"num_tokens": 368314.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.613600921630859,
|
|
"epoch": 0.16810540663334847,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.4354,
|
|
"mean_token_accuracy": 0.08428428322076797,
|
|
"num_tokens": 378406.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.429571628570557,
|
|
"epoch": 0.17264879600181735,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.439,
|
|
"mean_token_accuracy": 0.08039730936288833,
|
|
"num_tokens": 388912.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.273902130126952,
|
|
"epoch": 0.17719218537028622,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.4157,
|
|
"mean_token_accuracy": 0.08314879797399044,
|
|
"num_tokens": 399290.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.167303562164307,
|
|
"epoch": 0.18173557473875512,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.322,
|
|
"mean_token_accuracy": 0.08215823769569397,
|
|
"num_tokens": 409757.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.126931190490723,
|
|
"epoch": 0.186278964107224,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.4599,
|
|
"mean_token_accuracy": 0.08315562084317207,
|
|
"num_tokens": 420182.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.04471731185913,
|
|
"epoch": 0.19082235347569287,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.3635,
|
|
"mean_token_accuracy": 0.08014154359698296,
|
|
"num_tokens": 429897.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.973057556152344,
|
|
"epoch": 0.19536574284416175,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.2597,
|
|
"mean_token_accuracy": 0.08170034214854241,
|
|
"num_tokens": 441131.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.934465074539185,
|
|
"epoch": 0.19990913221263062,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.2285,
|
|
"mean_token_accuracy": 0.08460488840937615,
|
|
"num_tokens": 451894.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.865971183776855,
|
|
"epoch": 0.2044525215810995,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.1586,
|
|
"mean_token_accuracy": 0.08472679629921913,
|
|
"num_tokens": 462613.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.905964708328247,
|
|
"epoch": 0.20899591094956838,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.2551,
|
|
"mean_token_accuracy": 0.08738799616694451,
|
|
"num_tokens": 471606.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.820685958862304,
|
|
"epoch": 0.21353930031803725,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.1499,
|
|
"mean_token_accuracy": 0.08987567648291588,
|
|
"num_tokens": 481559.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.777825260162354,
|
|
"epoch": 0.21808268968650613,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.1686,
|
|
"mean_token_accuracy": 0.0859978199005127,
|
|
"num_tokens": 492766.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.835345315933227,
|
|
"epoch": 0.222626079054975,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.2285,
|
|
"mean_token_accuracy": 0.08684551119804382,
|
|
"num_tokens": 503357.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.6760852336883545,
|
|
"epoch": 0.22716946842344388,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.0637,
|
|
"mean_token_accuracy": 0.08950686678290368,
|
|
"num_tokens": 513098.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.730396175384522,
|
|
"epoch": 0.23171285779191278,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1422,
|
|
"mean_token_accuracy": 0.08494780957698822,
|
|
"num_tokens": 522807.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.82570514678955,
|
|
"epoch": 0.23625624716038165,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.1568,
|
|
"mean_token_accuracy": 0.0862432986497879,
|
|
"num_tokens": 532670.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.713788843154907,
|
|
"epoch": 0.24079963652885053,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.0973,
|
|
"mean_token_accuracy": 0.0868481919169426,
|
|
"num_tokens": 543647.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.765906810760498,
|
|
"epoch": 0.2453430258973194,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.112,
|
|
"mean_token_accuracy": 0.09636801481246948,
|
|
"num_tokens": 553047.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.660856294631958,
|
|
"epoch": 0.24988641526578828,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.1678,
|
|
"mean_token_accuracy": 0.09444689303636551,
|
|
"num_tokens": 563413.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.606683969497681,
|
|
"epoch": 0.25442980463425713,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.1218,
|
|
"mean_token_accuracy": 0.089845459908247,
|
|
"num_tokens": 573790.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.731978559494019,
|
|
"epoch": 0.25897319400272606,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.0439,
|
|
"mean_token_accuracy": 0.09121683463454247,
|
|
"num_tokens": 582958.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.521587800979614,
|
|
"epoch": 0.26351658337119493,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.9904,
|
|
"mean_token_accuracy": 0.09767817184329033,
|
|
"num_tokens": 592886.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.548699855804443,
|
|
"epoch": 0.2680599727396638,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.0399,
|
|
"mean_token_accuracy": 0.09062462002038955,
|
|
"num_tokens": 603692.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.651094913482666,
|
|
"epoch": 0.2726033621081327,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.0823,
|
|
"mean_token_accuracy": 0.0895950011909008,
|
|
"num_tokens": 613534.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.61688346862793,
|
|
"epoch": 0.27714675147660156,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.0274,
|
|
"mean_token_accuracy": 0.09576684832572938,
|
|
"num_tokens": 623862.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.62919430732727,
|
|
"epoch": 0.28169014084507044,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.1036,
|
|
"mean_token_accuracy": 0.09453180059790611,
|
|
"num_tokens": 633759.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.475176191329956,
|
|
"epoch": 0.2862335302135393,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.9859,
|
|
"mean_token_accuracy": 0.09430009499192238,
|
|
"num_tokens": 643553.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.528373908996582,
|
|
"epoch": 0.2907769195820082,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.9115,
|
|
"mean_token_accuracy": 0.09864105582237244,
|
|
"num_tokens": 653714.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.604555463790893,
|
|
"epoch": 0.29532030895047706,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.9306,
|
|
"mean_token_accuracy": 0.10158697441220284,
|
|
"num_tokens": 662817.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.483990573883057,
|
|
"epoch": 0.29986369831894594,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.9631,
|
|
"mean_token_accuracy": 0.0952290877699852,
|
|
"num_tokens": 672381.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.382349872589112,
|
|
"epoch": 0.3044070876874148,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.8338,
|
|
"mean_token_accuracy": 0.09701873064041137,
|
|
"num_tokens": 682525.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.496461296081543,
|
|
"epoch": 0.3089504770558837,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.9359,
|
|
"mean_token_accuracy": 0.09759294539690018,
|
|
"num_tokens": 691826.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.394898748397827,
|
|
"epoch": 0.31349386642435256,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.9089,
|
|
"mean_token_accuracy": 0.10296000391244889,
|
|
"num_tokens": 701319.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.4353029251098635,
|
|
"epoch": 0.31803725579282144,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.998,
|
|
"mean_token_accuracy": 0.0935439720749855,
|
|
"num_tokens": 710888.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.49090256690979,
|
|
"epoch": 0.3225806451612903,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.9328,
|
|
"mean_token_accuracy": 0.09594348818063736,
|
|
"num_tokens": 722516.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.475078916549682,
|
|
"epoch": 0.3271240345297592,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.9103,
|
|
"mean_token_accuracy": 0.09800132587552071,
|
|
"num_tokens": 733058.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.39355149269104,
|
|
"epoch": 0.33166742389822806,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.974,
|
|
"mean_token_accuracy": 0.09975532218813896,
|
|
"num_tokens": 744832.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.496864604949951,
|
|
"epoch": 0.33621081326669694,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.9025,
|
|
"mean_token_accuracy": 0.09925103634595871,
|
|
"num_tokens": 755387.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.3132157802581785,
|
|
"epoch": 0.3407542026351658,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.8349,
|
|
"mean_token_accuracy": 0.1016771987080574,
|
|
"num_tokens": 764455.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.339641284942627,
|
|
"epoch": 0.3452975920036347,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.8903,
|
|
"mean_token_accuracy": 0.09662013798952103,
|
|
"num_tokens": 774513.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.49444260597229,
|
|
"epoch": 0.34984098137210357,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.9771,
|
|
"mean_token_accuracy": 0.09587227553129196,
|
|
"num_tokens": 785540.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.336615896224975,
|
|
"epoch": 0.35438437074057244,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.8417,
|
|
"mean_token_accuracy": 0.09714982733130455,
|
|
"num_tokens": 795233.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.437350511550903,
|
|
"epoch": 0.35892776010904137,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.8975,
|
|
"mean_token_accuracy": 0.09597025141119957,
|
|
"num_tokens": 804855.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.3508072853088375,
|
|
"epoch": 0.36347114947751025,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.8581,
|
|
"mean_token_accuracy": 0.10273054391145706,
|
|
"num_tokens": 814818.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.307885694503784,
|
|
"epoch": 0.3680145388459791,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.8737,
|
|
"mean_token_accuracy": 0.10590235516428947,
|
|
"num_tokens": 824622.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.379908847808838,
|
|
"epoch": 0.372557928214448,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.7626,
|
|
"mean_token_accuracy": 0.10957324355840684,
|
|
"num_tokens": 834813.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.205720233917236,
|
|
"epoch": 0.3771013175829169,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.8537,
|
|
"mean_token_accuracy": 0.10128433629870415,
|
|
"num_tokens": 844887.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.279560470581055,
|
|
"epoch": 0.38164470695138575,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.8512,
|
|
"mean_token_accuracy": 0.10131347700953483,
|
|
"num_tokens": 855430.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.237056398391724,
|
|
"epoch": 0.3861880963198546,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.6942,
|
|
"mean_token_accuracy": 0.10520395040512084,
|
|
"num_tokens": 866052.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.226158237457275,
|
|
"epoch": 0.3907314856883235,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.8429,
|
|
"mean_token_accuracy": 0.0950412966310978,
|
|
"num_tokens": 877961.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.291323947906494,
|
|
"epoch": 0.3952748750567924,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.8103,
|
|
"mean_token_accuracy": 0.10648028403520585,
|
|
"num_tokens": 888386.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.235066747665405,
|
|
"epoch": 0.39981826442526125,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.8563,
|
|
"mean_token_accuracy": 0.10282102078199387,
|
|
"num_tokens": 898272.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.281696557998657,
|
|
"epoch": 0.4043616537937301,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.8014,
|
|
"mean_token_accuracy": 0.1076163724064827,
|
|
"num_tokens": 908317.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.248809957504273,
|
|
"epoch": 0.408905043162199,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.7819,
|
|
"mean_token_accuracy": 0.10194819942116737,
|
|
"num_tokens": 917970.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.240721273422241,
|
|
"epoch": 0.4134484325306679,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.8191,
|
|
"mean_token_accuracy": 0.09906565770506859,
|
|
"num_tokens": 928132.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.241307735443115,
|
|
"epoch": 0.41799182189913675,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.747,
|
|
"mean_token_accuracy": 0.1059770368039608,
|
|
"num_tokens": 938092.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.299763536453247,
|
|
"epoch": 0.4225352112676056,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.8651,
|
|
"mean_token_accuracy": 0.10198809579014778,
|
|
"num_tokens": 948650.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.130692863464356,
|
|
"epoch": 0.4270786006360745,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.7675,
|
|
"mean_token_accuracy": 0.10987066626548767,
|
|
"num_tokens": 960195.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.254830408096313,
|
|
"epoch": 0.4316219900045434,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.7984,
|
|
"mean_token_accuracy": 0.10485964193940163,
|
|
"num_tokens": 970608.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.314376783370972,
|
|
"epoch": 0.43616537937301225,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.7811,
|
|
"mean_token_accuracy": 0.10285131707787513,
|
|
"num_tokens": 981053.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.066157007217408,
|
|
"epoch": 0.4407087687414811,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.703,
|
|
"mean_token_accuracy": 0.11324564591050149,
|
|
"num_tokens": 991560.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.167983913421631,
|
|
"epoch": 0.44525215810995,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.7507,
|
|
"mean_token_accuracy": 0.1001538448035717,
|
|
"num_tokens": 1002008.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.145689487457275,
|
|
"epoch": 0.4497955474784189,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.7502,
|
|
"mean_token_accuracy": 0.10835718214511872,
|
|
"num_tokens": 1012349.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.15342149734497,
|
|
"epoch": 0.45433893684688775,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.7273,
|
|
"mean_token_accuracy": 0.10544943064451218,
|
|
"num_tokens": 1022886.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.45433893684688775,
|
|
"eval_entropy": 7.164822139296421,
|
|
"eval_loss": 6.852421760559082,
|
|
"eval_mean_token_accuracy": 0.10633610026954218,
|
|
"eval_num_tokens": 1022886.0,
|
|
"eval_runtime": 2.0057,
|
|
"eval_samples_per_second": 1712.63,
|
|
"eval_steps_per_second": 214.39,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.286551141738892,
|
|
"epoch": 0.45888232621535663,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.783,
|
|
"mean_token_accuracy": 0.09997680187225341,
|
|
"num_tokens": 1033129.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.133803033828736,
|
|
"epoch": 0.46342571558382556,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.6802,
|
|
"mean_token_accuracy": 0.10805964693427086,
|
|
"num_tokens": 1043069.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.079697799682617,
|
|
"epoch": 0.46796910495229443,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.6994,
|
|
"mean_token_accuracy": 0.1062344878911972,
|
|
"num_tokens": 1054049.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.0915602684021,
|
|
"epoch": 0.4725124943207633,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.6878,
|
|
"mean_token_accuracy": 0.1027360238134861,
|
|
"num_tokens": 1065155.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.116834926605224,
|
|
"epoch": 0.4770558836892322,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.6763,
|
|
"mean_token_accuracy": 0.1094151757657528,
|
|
"num_tokens": 1075510.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.024749040603638,
|
|
"epoch": 0.48159927305770106,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.6901,
|
|
"mean_token_accuracy": 0.1099021278321743,
|
|
"num_tokens": 1086494.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.2192497730255125,
|
|
"epoch": 0.48614266242616994,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.7271,
|
|
"mean_token_accuracy": 0.10860580801963807,
|
|
"num_tokens": 1097311.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.049293804168701,
|
|
"epoch": 0.4906860517946388,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.6835,
|
|
"mean_token_accuracy": 0.10724506676197051,
|
|
"num_tokens": 1107757.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.006491184234619,
|
|
"epoch": 0.4952294411631077,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.5917,
|
|
"mean_token_accuracy": 0.11055039316415786,
|
|
"num_tokens": 1118889.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.137886428833008,
|
|
"epoch": 0.49977283053157656,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.7913,
|
|
"mean_token_accuracy": 0.10480824410915375,
|
|
"num_tokens": 1128951.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.029620742797851,
|
|
"epoch": 0.5043162199000454,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.5948,
|
|
"mean_token_accuracy": 0.11510317772626877,
|
|
"num_tokens": 1138583.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.0846200466156,
|
|
"epoch": 0.5088596092685143,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.6631,
|
|
"mean_token_accuracy": 0.11140834540128708,
|
|
"num_tokens": 1149572.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.074535894393921,
|
|
"epoch": 0.5134029986369832,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.6367,
|
|
"mean_token_accuracy": 0.1094569742679596,
|
|
"num_tokens": 1160532.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.030823421478272,
|
|
"epoch": 0.5179463880054521,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.6299,
|
|
"mean_token_accuracy": 0.10860486924648285,
|
|
"num_tokens": 1170915.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.998789310455322,
|
|
"epoch": 0.522489777373921,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.5964,
|
|
"mean_token_accuracy": 0.11754036694765091,
|
|
"num_tokens": 1180822.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.999447011947632,
|
|
"epoch": 0.5270331667423899,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.6302,
|
|
"mean_token_accuracy": 0.11184115558862687,
|
|
"num_tokens": 1190447.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.026746368408203,
|
|
"epoch": 0.5315765561108587,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.6657,
|
|
"mean_token_accuracy": 0.11019867807626724,
|
|
"num_tokens": 1200951.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.064433956146241,
|
|
"epoch": 0.5361199454793276,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.5523,
|
|
"mean_token_accuracy": 0.1147008366882801,
|
|
"num_tokens": 1211846.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.766572427749634,
|
|
"epoch": 0.5406633348477965,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.5164,
|
|
"mean_token_accuracy": 0.11493531465530396,
|
|
"num_tokens": 1222074.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.925977277755737,
|
|
"epoch": 0.5452067242162654,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.5846,
|
|
"mean_token_accuracy": 0.11097392141819,
|
|
"num_tokens": 1231741.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.0018980503082275,
|
|
"epoch": 0.5497501135847342,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.5311,
|
|
"mean_token_accuracy": 0.11044337376952171,
|
|
"num_tokens": 1241078.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.9610405445098875,
|
|
"epoch": 0.5542935029532031,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.7226,
|
|
"mean_token_accuracy": 0.1040603518486023,
|
|
"num_tokens": 1251685.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.949973773956299,
|
|
"epoch": 0.558836892321672,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.562,
|
|
"mean_token_accuracy": 0.1197849079966545,
|
|
"num_tokens": 1262044.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.914613580703735,
|
|
"epoch": 0.5633802816901409,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.5713,
|
|
"mean_token_accuracy": 0.1111277848482132,
|
|
"num_tokens": 1273647.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.064757633209228,
|
|
"epoch": 0.5679236710586097,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.625,
|
|
"mean_token_accuracy": 0.10996143072843552,
|
|
"num_tokens": 1284396.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.925110483169556,
|
|
"epoch": 0.5724670604270786,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.6058,
|
|
"mean_token_accuracy": 0.1135744608938694,
|
|
"num_tokens": 1294339.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.936828088760376,
|
|
"epoch": 0.5770104497955475,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.524,
|
|
"mean_token_accuracy": 0.11486122459173202,
|
|
"num_tokens": 1304572.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.9446367740631105,
|
|
"epoch": 0.5815538391640164,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.5734,
|
|
"mean_token_accuracy": 0.11208348348736763,
|
|
"num_tokens": 1314780.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.862443494796753,
|
|
"epoch": 0.5860972285324852,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.5789,
|
|
"mean_token_accuracy": 0.11142232716083526,
|
|
"num_tokens": 1325215.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.912387943267822,
|
|
"epoch": 0.5906406179009541,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.4907,
|
|
"mean_token_accuracy": 0.11991971731185913,
|
|
"num_tokens": 1334745.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.957016134262085,
|
|
"epoch": 0.595184007269423,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.6005,
|
|
"mean_token_accuracy": 0.11654155552387238,
|
|
"num_tokens": 1345150.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.955321359634399,
|
|
"epoch": 0.5997273966378919,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.554,
|
|
"mean_token_accuracy": 0.1150270789861679,
|
|
"num_tokens": 1355619.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.841307306289673,
|
|
"epoch": 0.6042707860063607,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.4846,
|
|
"mean_token_accuracy": 0.11608346402645112,
|
|
"num_tokens": 1365933.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.891475486755371,
|
|
"epoch": 0.6088141753748296,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.5081,
|
|
"mean_token_accuracy": 0.11709481626749038,
|
|
"num_tokens": 1376867.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.965159177780151,
|
|
"epoch": 0.6133575647432985,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.6588,
|
|
"mean_token_accuracy": 0.1138373926281929,
|
|
"num_tokens": 1386739.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.889486598968506,
|
|
"epoch": 0.6179009541117674,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.5181,
|
|
"mean_token_accuracy": 0.116570845246315,
|
|
"num_tokens": 1397138.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.962109041213989,
|
|
"epoch": 0.6224443434802363,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.6159,
|
|
"mean_token_accuracy": 0.11419135332107544,
|
|
"num_tokens": 1408171.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.836606550216675,
|
|
"epoch": 0.6269877328487051,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.5509,
|
|
"mean_token_accuracy": 0.10746546238660812,
|
|
"num_tokens": 1418259.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.8779538631439205,
|
|
"epoch": 0.631531122217174,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.5466,
|
|
"mean_token_accuracy": 0.1198190651834011,
|
|
"num_tokens": 1428080.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.9482708930969235,
|
|
"epoch": 0.6360745115856429,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.5777,
|
|
"mean_token_accuracy": 0.11636564508080482,
|
|
"num_tokens": 1438502.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.803214025497437,
|
|
"epoch": 0.6406179009541118,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.5271,
|
|
"mean_token_accuracy": 0.11418517231941223,
|
|
"num_tokens": 1448396.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.889286518096924,
|
|
"epoch": 0.6451612903225806,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.581,
|
|
"mean_token_accuracy": 0.10854673013091087,
|
|
"num_tokens": 1459728.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.843706035614014,
|
|
"epoch": 0.6497046796910495,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.4748,
|
|
"mean_token_accuracy": 0.12437612041831017,
|
|
"num_tokens": 1469733.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.759375,
|
|
"epoch": 0.6542480690595184,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4664,
|
|
"mean_token_accuracy": 0.1201399251818657,
|
|
"num_tokens": 1479004.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.855831241607666,
|
|
"epoch": 0.6587914584279873,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.4365,
|
|
"mean_token_accuracy": 0.11637238636612893,
|
|
"num_tokens": 1489848.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.7456646919250485,
|
|
"epoch": 0.6633348477964561,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.4363,
|
|
"mean_token_accuracy": 0.1179615817964077,
|
|
"num_tokens": 1500877.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.931111192703247,
|
|
"epoch": 0.667878237164925,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.5786,
|
|
"mean_token_accuracy": 0.11474068090319633,
|
|
"num_tokens": 1511840.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.754684495925903,
|
|
"epoch": 0.6724216265333939,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.4844,
|
|
"mean_token_accuracy": 0.11675599068403245,
|
|
"num_tokens": 1521839.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.837312364578247,
|
|
"epoch": 0.6769650159018628,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.3889,
|
|
"mean_token_accuracy": 0.12173137590289115,
|
|
"num_tokens": 1531285.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.607749176025391,
|
|
"epoch": 0.6815084052703316,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.4363,
|
|
"mean_token_accuracy": 0.11859396696090699,
|
|
"num_tokens": 1542027.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.8664957046508786,
|
|
"epoch": 0.6860517946388005,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.548,
|
|
"mean_token_accuracy": 0.11864770278334617,
|
|
"num_tokens": 1552527.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.915476655960083,
|
|
"epoch": 0.6905951840072694,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.5848,
|
|
"mean_token_accuracy": 0.11684316471219063,
|
|
"num_tokens": 1562605.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.740834951400757,
|
|
"epoch": 0.6951385733757383,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.4686,
|
|
"mean_token_accuracy": 0.11819475442171097,
|
|
"num_tokens": 1572128.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.718352746963501,
|
|
"epoch": 0.6996819627442071,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.4003,
|
|
"mean_token_accuracy": 0.11825122833251953,
|
|
"num_tokens": 1582352.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.808340549468994,
|
|
"epoch": 0.704225352112676,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.4936,
|
|
"mean_token_accuracy": 0.12305623516440392,
|
|
"num_tokens": 1592887.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.696469831466675,
|
|
"epoch": 0.7087687414811449,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.4367,
|
|
"mean_token_accuracy": 0.11979465782642365,
|
|
"num_tokens": 1603564.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.64963583946228,
|
|
"epoch": 0.7133121308496138,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.2928,
|
|
"mean_token_accuracy": 0.12856932654976844,
|
|
"num_tokens": 1613270.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.67563099861145,
|
|
"epoch": 0.7178555202180827,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.3618,
|
|
"mean_token_accuracy": 0.12818560600280762,
|
|
"num_tokens": 1623009.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.701826286315918,
|
|
"epoch": 0.7223989095865516,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.3966,
|
|
"mean_token_accuracy": 0.11920078694820405,
|
|
"num_tokens": 1633209.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.767777013778686,
|
|
"epoch": 0.7269422989550205,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.4338,
|
|
"mean_token_accuracy": 0.12207518145442009,
|
|
"num_tokens": 1643708.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.734032678604126,
|
|
"epoch": 0.7314856883234894,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.4948,
|
|
"mean_token_accuracy": 0.11701320484280586,
|
|
"num_tokens": 1653973.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.704824733734131,
|
|
"epoch": 0.7360290776919582,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.4747,
|
|
"mean_token_accuracy": 0.12068637907505035,
|
|
"num_tokens": 1664856.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.683869314193726,
|
|
"epoch": 0.7405724670604271,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.3944,
|
|
"mean_token_accuracy": 0.12224291115999222,
|
|
"num_tokens": 1675935.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.7558623313903805,
|
|
"epoch": 0.745115856428896,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.4428,
|
|
"mean_token_accuracy": 0.12248818576335907,
|
|
"num_tokens": 1686597.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.610339832305908,
|
|
"epoch": 0.7496592457973649,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.3911,
|
|
"mean_token_accuracy": 0.1245286226272583,
|
|
"num_tokens": 1697385.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.8567352294921875,
|
|
"epoch": 0.7542026351658337,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.4443,
|
|
"mean_token_accuracy": 0.11997077614068985,
|
|
"num_tokens": 1706680.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.645043087005615,
|
|
"epoch": 0.7587460245343026,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.4407,
|
|
"mean_token_accuracy": 0.1258401207625866,
|
|
"num_tokens": 1716130.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.69645881652832,
|
|
"epoch": 0.7632894139027715,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.3624,
|
|
"mean_token_accuracy": 0.11782209128141403,
|
|
"num_tokens": 1726835.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.711601686477661,
|
|
"epoch": 0.7678328032712404,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.3928,
|
|
"mean_token_accuracy": 0.1290561191737652,
|
|
"num_tokens": 1737708.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.639672756195068,
|
|
"epoch": 0.7723761926397092,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.3799,
|
|
"mean_token_accuracy": 0.12615877538919448,
|
|
"num_tokens": 1747649.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.673767709732056,
|
|
"epoch": 0.7769195820081781,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.4068,
|
|
"mean_token_accuracy": 0.1206041231751442,
|
|
"num_tokens": 1758327.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.613749980926514,
|
|
"epoch": 0.781462971376647,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.3619,
|
|
"mean_token_accuracy": 0.121267169713974,
|
|
"num_tokens": 1769767.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.713872861862183,
|
|
"epoch": 0.7860063607451159,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.4717,
|
|
"mean_token_accuracy": 0.1220657117664814,
|
|
"num_tokens": 1780139.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.726572799682617,
|
|
"epoch": 0.7905497501135847,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.3601,
|
|
"mean_token_accuracy": 0.12505988776683807,
|
|
"num_tokens": 1790783.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.59320387840271,
|
|
"epoch": 0.7950931394820536,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.3513,
|
|
"mean_token_accuracy": 0.12245700359344483,
|
|
"num_tokens": 1801349.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.600114583969116,
|
|
"epoch": 0.7996365288505225,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.3531,
|
|
"mean_token_accuracy": 0.1229885309934616,
|
|
"num_tokens": 1811990.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.57788610458374,
|
|
"epoch": 0.8041799182189914,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.3634,
|
|
"mean_token_accuracy": 0.13060024604201317,
|
|
"num_tokens": 1821725.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.60519003868103,
|
|
"epoch": 0.8087233075874602,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.3253,
|
|
"mean_token_accuracy": 0.11931257545948029,
|
|
"num_tokens": 1831794.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.64742259979248,
|
|
"epoch": 0.8132666969559291,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.4077,
|
|
"mean_token_accuracy": 0.12239367365837098,
|
|
"num_tokens": 1842002.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.655037879943848,
|
|
"epoch": 0.817810086324398,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.3797,
|
|
"mean_token_accuracy": 0.12452054545283317,
|
|
"num_tokens": 1852282.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.614450359344483,
|
|
"epoch": 0.8223534756928669,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.388,
|
|
"mean_token_accuracy": 0.11854704096913338,
|
|
"num_tokens": 1862830.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.592864656448365,
|
|
"epoch": 0.8268968650613358,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.3413,
|
|
"mean_token_accuracy": 0.1236710526049137,
|
|
"num_tokens": 1872746.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.5772138118743895,
|
|
"epoch": 0.8314402544298046,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.3719,
|
|
"mean_token_accuracy": 0.11749232411384583,
|
|
"num_tokens": 1884353.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.676659631729126,
|
|
"epoch": 0.8359836437982735,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.4017,
|
|
"mean_token_accuracy": 0.11830696910619735,
|
|
"num_tokens": 1895719.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.629153537750244,
|
|
"epoch": 0.8405270331667424,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.3444,
|
|
"mean_token_accuracy": 0.13160407915711403,
|
|
"num_tokens": 1905905.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.659778261184693,
|
|
"epoch": 0.8450704225352113,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.4383,
|
|
"mean_token_accuracy": 0.12295959070324898,
|
|
"num_tokens": 1916010.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.56156849861145,
|
|
"epoch": 0.8496138119036801,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.3334,
|
|
"mean_token_accuracy": 0.12297626435756684,
|
|
"num_tokens": 1926429.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.445285606384277,
|
|
"epoch": 0.854157201272149,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.2646,
|
|
"mean_token_accuracy": 0.12663295194506646,
|
|
"num_tokens": 1936322.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.68843822479248,
|
|
"epoch": 0.8587005906406179,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.4278,
|
|
"mean_token_accuracy": 0.12256157100200653,
|
|
"num_tokens": 1947885.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.635422468185425,
|
|
"epoch": 0.8632439800090868,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.3923,
|
|
"mean_token_accuracy": 0.12212013304233552,
|
|
"num_tokens": 1957876.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.575307750701905,
|
|
"epoch": 0.8677873693775556,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.3884,
|
|
"mean_token_accuracy": 0.11944693848490714,
|
|
"num_tokens": 1969531.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.577566289901734,
|
|
"epoch": 0.8723307587460245,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.2501,
|
|
"mean_token_accuracy": 0.12427858263254166,
|
|
"num_tokens": 1980787.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.506700944900513,
|
|
"epoch": 0.8768741481144934,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.3301,
|
|
"mean_token_accuracy": 0.12719210535287856,
|
|
"num_tokens": 1991645.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.563048219680786,
|
|
"epoch": 0.8814175374829623,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.38,
|
|
"mean_token_accuracy": 0.12680814787745476,
|
|
"num_tokens": 2002544.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.701274681091308,
|
|
"epoch": 0.8859609268514311,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.4409,
|
|
"mean_token_accuracy": 0.12047607451677322,
|
|
"num_tokens": 2012813.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.542685174942017,
|
|
"epoch": 0.8905043162199,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.3017,
|
|
"mean_token_accuracy": 0.1271137699484825,
|
|
"num_tokens": 2023756.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.471768379211426,
|
|
"epoch": 0.8950477055883689,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.3821,
|
|
"mean_token_accuracy": 0.12684743776917456,
|
|
"num_tokens": 2034245.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.698192596435547,
|
|
"epoch": 0.8995910949568378,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.4087,
|
|
"mean_token_accuracy": 0.12129435613751412,
|
|
"num_tokens": 2045635.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.564974308013916,
|
|
"epoch": 0.9041344843253066,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.3742,
|
|
"mean_token_accuracy": 0.12397633790969849,
|
|
"num_tokens": 2056501.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.569272136688232,
|
|
"epoch": 0.9086778736937755,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.3834,
|
|
"mean_token_accuracy": 0.12595843970775605,
|
|
"num_tokens": 2066540.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.9086778736937755,
|
|
"eval_entropy": 6.449487343499827,
|
|
"eval_loss": 6.4225172996521,
|
|
"eval_mean_token_accuracy": 0.12387349221768768,
|
|
"eval_num_tokens": 2066540.0,
|
|
"eval_runtime": 1.9962,
|
|
"eval_samples_per_second": 1720.779,
|
|
"eval_steps_per_second": 215.41,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.589289760589599,
|
|
"epoch": 0.9132212630622444,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999998223471442,
|
|
"loss": 6.27,
|
|
"mean_token_accuracy": 0.12683147192001343,
|
|
"num_tokens": 2077008.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.468321657180786,
|
|
"epoch": 0.9177646524307133,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999991006328981,
|
|
"loss": 6.1974,
|
|
"mean_token_accuracy": 0.12910854443907738,
|
|
"num_tokens": 2087314.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.515331792831421,
|
|
"epoch": 0.9223080417991822,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999978237557377,
|
|
"loss": 6.365,
|
|
"mean_token_accuracy": 0.1273949436843395,
|
|
"num_tokens": 2097928.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.527489137649536,
|
|
"epoch": 0.9268514311676511,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999959917188137,
|
|
"loss": 6.2922,
|
|
"mean_token_accuracy": 0.12393052130937576,
|
|
"num_tokens": 2108662.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.452746248245239,
|
|
"epoch": 0.93139482053612,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999936045266462,
|
|
"loss": 6.3291,
|
|
"mean_token_accuracy": 0.12396076545119286,
|
|
"num_tokens": 2118546.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.609079790115357,
|
|
"epoch": 0.9359382099045889,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999906621851255,
|
|
"loss": 6.3514,
|
|
"mean_token_accuracy": 0.1282046727836132,
|
|
"num_tokens": 2128012.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.469990539550781,
|
|
"epoch": 0.9404815992730577,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999871647015115,
|
|
"loss": 6.1834,
|
|
"mean_token_accuracy": 0.12837918177247049,
|
|
"num_tokens": 2138241.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.617193222045898,
|
|
"epoch": 0.9450249886415266,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999831120844339,
|
|
"loss": 6.3515,
|
|
"mean_token_accuracy": 0.1252776153385639,
|
|
"num_tokens": 2150780.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.467695426940918,
|
|
"epoch": 0.9495683780099955,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999785043438922,
|
|
"loss": 6.2353,
|
|
"mean_token_accuracy": 0.12729288563132285,
|
|
"num_tokens": 2161918.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.440260601043701,
|
|
"epoch": 0.9541117673784644,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999733414912555,
|
|
"loss": 6.2691,
|
|
"mean_token_accuracy": 0.12473016008734703,
|
|
"num_tokens": 2172742.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.444692325592041,
|
|
"epoch": 0.9586551567469332,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999676235392626,
|
|
"loss": 6.2628,
|
|
"mean_token_accuracy": 0.1319998823106289,
|
|
"num_tokens": 2182122.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.485652875900269,
|
|
"epoch": 0.9631985461154021,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999613505020219,
|
|
"loss": 6.3009,
|
|
"mean_token_accuracy": 0.12624281868338585,
|
|
"num_tokens": 2193703.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.477052927017212,
|
|
"epoch": 0.967741935483871,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999545223950116,
|
|
"loss": 6.232,
|
|
"mean_token_accuracy": 0.12531697526574134,
|
|
"num_tokens": 2204041.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.4596601009368895,
|
|
"epoch": 0.9722853248523399,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999471392350795,
|
|
"loss": 6.2681,
|
|
"mean_token_accuracy": 0.1249091774225235,
|
|
"num_tokens": 2214276.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.514900493621826,
|
|
"epoch": 0.9768287142208087,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999392010404426,
|
|
"loss": 6.237,
|
|
"mean_token_accuracy": 0.13163011744618416,
|
|
"num_tokens": 2225390.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.385442638397217,
|
|
"epoch": 0.9813721035892776,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999307078306876,
|
|
"loss": 6.1923,
|
|
"mean_token_accuracy": 0.13028257265686988,
|
|
"num_tokens": 2235708.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.4769080638885494,
|
|
"epoch": 0.9859154929577465,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999216596267708,
|
|
"loss": 6.2303,
|
|
"mean_token_accuracy": 0.13035036474466324,
|
|
"num_tokens": 2246284.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.524853515625,
|
|
"epoch": 0.9904588823262154,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999120564510177,
|
|
"loss": 6.2358,
|
|
"mean_token_accuracy": 0.12543209120631219,
|
|
"num_tokens": 2257219.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.371940565109253,
|
|
"epoch": 0.9950022716946842,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000499901898327123,
|
|
"loss": 6.2336,
|
|
"mean_token_accuracy": 0.12906517162919046,
|
|
"num_tokens": 2268019.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.449350881576538,
|
|
"epoch": 0.9995456610631531,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004998911852801512,
|
|
"loss": 6.253,
|
|
"mean_token_accuracy": 0.12894620150327682,
|
|
"num_tokens": 2278382.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.441971778869629,
|
|
"epoch": 1.003634711494775,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004998799173365355,
|
|
"loss": 6.096,
|
|
"mean_token_accuracy": 0.1329502835869789,
|
|
"num_tokens": 2288515.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.444106578826904,
|
|
"epoch": 1.008178100863244,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998680945240783,
|
|
"loss": 5.9544,
|
|
"mean_token_accuracy": 0.1410957932472229,
|
|
"num_tokens": 2298396.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.330610656738282,
|
|
"epoch": 1.0127214902317128,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998557168719513,
|
|
"loss": 6.0684,
|
|
"mean_token_accuracy": 0.12948169335722923,
|
|
"num_tokens": 2308707.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.406459140777588,
|
|
"epoch": 1.0172648796001817,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998427844106954,
|
|
"loss": 5.98,
|
|
"mean_token_accuracy": 0.14317394495010377,
|
|
"num_tokens": 2318981.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.371292638778686,
|
|
"epoch": 1.0218082689686505,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998292971722197,
|
|
"loss": 6.072,
|
|
"mean_token_accuracy": 0.13840934932231902,
|
|
"num_tokens": 2329647.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.340481090545654,
|
|
"epoch": 1.0263516583371195,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499815255189803,
|
|
"loss": 6.0067,
|
|
"mean_token_accuracy": 0.13939314261078833,
|
|
"num_tokens": 2339449.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.3922374725341795,
|
|
"epoch": 1.0308950477055883,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998006584980922,
|
|
"loss": 6.1137,
|
|
"mean_token_accuracy": 0.13605127185583116,
|
|
"num_tokens": 2349904.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.3493505954742435,
|
|
"epoch": 1.0354384370740572,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004997855071331034,
|
|
"loss": 6.0212,
|
|
"mean_token_accuracy": 0.14467629939317703,
|
|
"num_tokens": 2359833.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.463475131988526,
|
|
"epoch": 1.039981826442526,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997698011322211,
|
|
"loss": 6.0567,
|
|
"mean_token_accuracy": 0.13746307641267777,
|
|
"num_tokens": 2369857.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.399654150009155,
|
|
"epoch": 1.044525215810995,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000499753540534198,
|
|
"loss": 6.0661,
|
|
"mean_token_accuracy": 0.14009104669094086,
|
|
"num_tokens": 2380372.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.322651815414429,
|
|
"epoch": 1.049068605179464,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000499736725379156,
|
|
"loss": 6.046,
|
|
"mean_token_accuracy": 0.13437431380152703,
|
|
"num_tokens": 2391239.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.3150145530700685,
|
|
"epoch": 1.0536119945479328,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004997193557085844,
|
|
"loss": 6.0224,
|
|
"mean_token_accuracy": 0.14738521873950958,
|
|
"num_tokens": 2401178.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.440890884399414,
|
|
"epoch": 1.0581553839164017,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004997014315653413,
|
|
"loss": 6.115,
|
|
"mean_token_accuracy": 0.13918990567326545,
|
|
"num_tokens": 2412496.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.359582996368408,
|
|
"epoch": 1.0626987732848705,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996829529936527,
|
|
"loss": 6.0976,
|
|
"mean_token_accuracy": 0.1375906229019165,
|
|
"num_tokens": 2423281.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.327804279327393,
|
|
"epoch": 1.0672421626533395,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004996639200391126,
|
|
"loss": 6.03,
|
|
"mean_token_accuracy": 0.14599778205156327,
|
|
"num_tokens": 2434078.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.311521911621094,
|
|
"epoch": 1.0717855520218083,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004996443327486831,
|
|
"loss": 5.9602,
|
|
"mean_token_accuracy": 0.141278412938118,
|
|
"num_tokens": 2443808.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.14685435295105,
|
|
"epoch": 1.0763289413902772,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996241911706937,
|
|
"loss": 6.0035,
|
|
"mean_token_accuracy": 0.14571388214826583,
|
|
"num_tokens": 2453397.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.454845142364502,
|
|
"epoch": 1.080872330758746,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004996034953548418,
|
|
"loss": 6.1026,
|
|
"mean_token_accuracy": 0.13732529804110527,
|
|
"num_tokens": 2464295.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.430036211013794,
|
|
"epoch": 1.085415720127215,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004995822453521924,
|
|
"loss": 6.1424,
|
|
"mean_token_accuracy": 0.13445171639323233,
|
|
"num_tokens": 2475311.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.353321695327759,
|
|
"epoch": 1.0899591094956838,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004995604412151776,
|
|
"loss": 6.0502,
|
|
"mean_token_accuracy": 0.13472085893154145,
|
|
"num_tokens": 2485968.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.279279947280884,
|
|
"epoch": 1.0945024988641527,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000499538082997597,
|
|
"loss": 6.0794,
|
|
"mean_token_accuracy": 0.13565608412027358,
|
|
"num_tokens": 2496183.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.318364000320434,
|
|
"epoch": 1.0990458882326215,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004995151707546172,
|
|
"loss": 6.0395,
|
|
"mean_token_accuracy": 0.13798854872584343,
|
|
"num_tokens": 2506242.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.372706079483033,
|
|
"epoch": 1.1035892776010905,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004994917045427721,
|
|
"loss": 6.0423,
|
|
"mean_token_accuracy": 0.14290895834565162,
|
|
"num_tokens": 2516443.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.290970277786255,
|
|
"epoch": 1.1081326669695593,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004994676844199621,
|
|
"loss": 6.0137,
|
|
"mean_token_accuracy": 0.1438373804092407,
|
|
"num_tokens": 2525795.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.2475104331970215,
|
|
"epoch": 1.1126760563380282,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004994431104454543,
|
|
"loss": 6.0515,
|
|
"mean_token_accuracy": 0.13550007790327073,
|
|
"num_tokens": 2535046.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.374474906921387,
|
|
"epoch": 1.117219445706497,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004994179826798828,
|
|
"loss": 6.0319,
|
|
"mean_token_accuracy": 0.1388425625860691,
|
|
"num_tokens": 2546489.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.1246405124664305,
|
|
"epoch": 1.121762835074966,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004993923011852477,
|
|
"loss": 5.9768,
|
|
"mean_token_accuracy": 0.14362312108278275,
|
|
"num_tokens": 2556335.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.372661924362182,
|
|
"epoch": 1.1263062244434348,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004993660660249158,
|
|
"loss": 6.0025,
|
|
"mean_token_accuracy": 0.13659965619444847,
|
|
"num_tokens": 2567647.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.418256092071533,
|
|
"epoch": 1.1308496138119037,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004993392772636194,
|
|
"loss": 6.1952,
|
|
"mean_token_accuracy": 0.13226437345147132,
|
|
"num_tokens": 2578618.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.217349433898926,
|
|
"epoch": 1.1353930031803725,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004993119349674573,
|
|
"loss": 6.0181,
|
|
"mean_token_accuracy": 0.14242995530366898,
|
|
"num_tokens": 2587797.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.2919868469238285,
|
|
"epoch": 1.1399363925488415,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004992840392038938,
|
|
"loss": 6.0506,
|
|
"mean_token_accuracy": 0.1346188299357891,
|
|
"num_tokens": 2598403.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.343607950210571,
|
|
"epoch": 1.1444797819173103,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499255590041759,
|
|
"loss": 6.0561,
|
|
"mean_token_accuracy": 0.13540524691343309,
|
|
"num_tokens": 2608296.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.313466644287109,
|
|
"epoch": 1.1490231712857792,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004992265875512485,
|
|
"loss": 6.0641,
|
|
"mean_token_accuracy": 0.14103504940867423,
|
|
"num_tokens": 2617906.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.25914363861084,
|
|
"epoch": 1.153566560654248,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004991970318039228,
|
|
"loss": 6.0032,
|
|
"mean_token_accuracy": 0.14189390316605568,
|
|
"num_tokens": 2627699.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.326404142379761,
|
|
"epoch": 1.158109950022717,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004991669228727078,
|
|
"loss": 5.9772,
|
|
"mean_token_accuracy": 0.14280295968055726,
|
|
"num_tokens": 2638384.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.2005486488342285,
|
|
"epoch": 1.1626533393911858,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004991362608318947,
|
|
"loss": 5.9212,
|
|
"mean_token_accuracy": 0.1442661963403225,
|
|
"num_tokens": 2648279.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.234303903579712,
|
|
"epoch": 1.1671967287596547,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004991050457571384,
|
|
"loss": 6.0072,
|
|
"mean_token_accuracy": 0.12986308410763742,
|
|
"num_tokens": 2657000.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.285914278030395,
|
|
"epoch": 1.1717401181281235,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004990732777254594,
|
|
"loss": 6.0186,
|
|
"mean_token_accuracy": 0.1422954961657524,
|
|
"num_tokens": 2667760.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.256361961364746,
|
|
"epoch": 1.1762835074965925,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004990409568152422,
|
|
"loss": 5.9631,
|
|
"mean_token_accuracy": 0.14491396695375441,
|
|
"num_tokens": 2678450.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.285534238815307,
|
|
"epoch": 1.1808268968650613,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004990080831062352,
|
|
"loss": 6.0766,
|
|
"mean_token_accuracy": 0.1332152709364891,
|
|
"num_tokens": 2689695.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.358760118484497,
|
|
"epoch": 1.1853702862335302,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004989746566795512,
|
|
"loss": 6.083,
|
|
"mean_token_accuracy": 0.14606243073940278,
|
|
"num_tokens": 2700414.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.18422908782959,
|
|
"epoch": 1.189913675601999,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004989406776176665,
|
|
"loss": 5.9112,
|
|
"mean_token_accuracy": 0.1534299984574318,
|
|
"num_tokens": 2709985.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.229423856735229,
|
|
"epoch": 1.194457064970468,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000498906146004421,
|
|
"loss": 6.0529,
|
|
"mean_token_accuracy": 0.14144230335950853,
|
|
"num_tokens": 2720990.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.253258466720581,
|
|
"epoch": 1.1990004543389368,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004988710619250181,
|
|
"loss": 5.9568,
|
|
"mean_token_accuracy": 0.14692321121692659,
|
|
"num_tokens": 2731285.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.206444644927979,
|
|
"epoch": 1.2035438437074057,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004988354254660244,
|
|
"loss": 6.0309,
|
|
"mean_token_accuracy": 0.13927505165338516,
|
|
"num_tokens": 2741024.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.2238139629364015,
|
|
"epoch": 1.2080872330758745,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004987992367153691,
|
|
"loss": 5.9787,
|
|
"mean_token_accuracy": 0.1435750648379326,
|
|
"num_tokens": 2751447.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.272238683700562,
|
|
"epoch": 1.2126306224443435,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004987624957623443,
|
|
"loss": 5.9818,
|
|
"mean_token_accuracy": 0.1429169222712517,
|
|
"num_tokens": 2761283.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.119753646850586,
|
|
"epoch": 1.2171740118128123,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004987252026976049,
|
|
"loss": 5.9119,
|
|
"mean_token_accuracy": 0.1478346861898899,
|
|
"num_tokens": 2771354.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.369901561737061,
|
|
"epoch": 1.2217174011812812,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004986873576131676,
|
|
"loss": 6.1054,
|
|
"mean_token_accuracy": 0.1408141702413559,
|
|
"num_tokens": 2781063.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.245864677429199,
|
|
"epoch": 1.22626079054975,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004986489606024116,
|
|
"loss": 6.008,
|
|
"mean_token_accuracy": 0.143255265802145,
|
|
"num_tokens": 2791103.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.304475927352906,
|
|
"epoch": 1.230804179918219,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004986100117600776,
|
|
"loss": 6.1001,
|
|
"mean_token_accuracy": 0.13890691325068474,
|
|
"num_tokens": 2802382.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.228710794448853,
|
|
"epoch": 1.235347569286688,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000498570511182268,
|
|
"loss": 5.9751,
|
|
"mean_token_accuracy": 0.14123353064060212,
|
|
"num_tokens": 2813740.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.206845951080322,
|
|
"epoch": 1.2398909586551567,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004985304589664465,
|
|
"loss": 5.9915,
|
|
"mean_token_accuracy": 0.13771087154746056,
|
|
"num_tokens": 2824121.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.323585081100464,
|
|
"epoch": 1.2444343480236255,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004984898552114381,
|
|
"loss": 6.0065,
|
|
"mean_token_accuracy": 0.1431947186589241,
|
|
"num_tokens": 2833356.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.192498970031738,
|
|
"epoch": 1.2489777373920945,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004984487000174283,
|
|
"loss": 5.9838,
|
|
"mean_token_accuracy": 0.13862482085824013,
|
|
"num_tokens": 2843839.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.203368473052978,
|
|
"epoch": 1.2535211267605635,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004984069934859637,
|
|
"loss": 5.9454,
|
|
"mean_token_accuracy": 0.1433752179145813,
|
|
"num_tokens": 2853523.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.2640033721923825,
|
|
"epoch": 1.2580645161290323,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004983647357199509,
|
|
"loss": 5.9947,
|
|
"mean_token_accuracy": 0.14069052115082742,
|
|
"num_tokens": 2863226.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.273294878005982,
|
|
"epoch": 1.262607905497501,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004983219268236567,
|
|
"loss": 6.0546,
|
|
"mean_token_accuracy": 0.1447424255311489,
|
|
"num_tokens": 2873822.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.226011180877686,
|
|
"epoch": 1.26715129486597,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004982785669027079,
|
|
"loss": 5.9184,
|
|
"mean_token_accuracy": 0.14830858260393143,
|
|
"num_tokens": 2883645.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.190831136703491,
|
|
"epoch": 1.271694684234439,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004982346560640907,
|
|
"loss": 6.033,
|
|
"mean_token_accuracy": 0.13505668118596076,
|
|
"num_tokens": 2894397.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.299043893814087,
|
|
"epoch": 1.2762380736029078,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004981901944161507,
|
|
"loss": 6.0134,
|
|
"mean_token_accuracy": 0.1432926282286644,
|
|
"num_tokens": 2904947.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.1759058952331545,
|
|
"epoch": 1.2807814629713765,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004981451820685928,
|
|
"loss": 5.9603,
|
|
"mean_token_accuracy": 0.1484628587961197,
|
|
"num_tokens": 2916183.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.175095415115356,
|
|
"epoch": 1.2853248523398455,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004980996191324802,
|
|
"loss": 5.8874,
|
|
"mean_token_accuracy": 0.14141177833080293,
|
|
"num_tokens": 2925429.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.246431970596314,
|
|
"epoch": 1.2898682417083145,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004980535057202351,
|
|
"loss": 5.9486,
|
|
"mean_token_accuracy": 0.1475185953080654,
|
|
"num_tokens": 2935142.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.173218536376953,
|
|
"epoch": 1.2944116310767833,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004980068419456378,
|
|
"loss": 6.0054,
|
|
"mean_token_accuracy": 0.1413328766822815,
|
|
"num_tokens": 2945151.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.3213526725769045,
|
|
"epoch": 1.2989550204452522,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004979596279238265,
|
|
"loss": 6.0962,
|
|
"mean_token_accuracy": 0.14202187061309815,
|
|
"num_tokens": 2956015.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.199492073059082,
|
|
"epoch": 1.303498409813721,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004979118637712969,
|
|
"loss": 5.9377,
|
|
"mean_token_accuracy": 0.14465177059173584,
|
|
"num_tokens": 2965879.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.216682767868042,
|
|
"epoch": 1.30804179918219,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004978635496059025,
|
|
"loss": 6.0119,
|
|
"mean_token_accuracy": 0.1414727032184601,
|
|
"num_tokens": 2977021.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.1556864261627195,
|
|
"epoch": 1.3125851885506588,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004978146855468537,
|
|
"loss": 5.9086,
|
|
"mean_token_accuracy": 0.14936657398939132,
|
|
"num_tokens": 2987169.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.278210592269898,
|
|
"epoch": 1.3171285779191277,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004977652717147177,
|
|
"loss": 6.0199,
|
|
"mean_token_accuracy": 0.14351356476545335,
|
|
"num_tokens": 2997666.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.191931247711182,
|
|
"epoch": 1.3216719672875965,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004977153082314182,
|
|
"loss": 5.8929,
|
|
"mean_token_accuracy": 0.14053303152322769,
|
|
"num_tokens": 3007021.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.170299625396728,
|
|
"epoch": 1.3262153566560655,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000497664795220235,
|
|
"loss": 5.952,
|
|
"mean_token_accuracy": 0.15345703661441804,
|
|
"num_tokens": 3016557.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.152959823608398,
|
|
"epoch": 1.3307587460245343,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004976137328058042,
|
|
"loss": 5.9222,
|
|
"mean_token_accuracy": 0.1459203690290451,
|
|
"num_tokens": 3026146.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.264169502258301,
|
|
"epoch": 1.3353021353930032,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000497562121114117,
|
|
"loss": 5.9734,
|
|
"mean_token_accuracy": 0.14671601578593255,
|
|
"num_tokens": 3037234.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.145901250839233,
|
|
"epoch": 1.339845524761472,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004975099602725202,
|
|
"loss": 5.9171,
|
|
"mean_token_accuracy": 0.14811089187860488,
|
|
"num_tokens": 3047282.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.0958422183990475,
|
|
"epoch": 1.344388914129941,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004974572504097154,
|
|
"loss": 5.9315,
|
|
"mean_token_accuracy": 0.14823032170534134,
|
|
"num_tokens": 3058648.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.407826995849609,
|
|
"epoch": 1.3489323034984098,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004974039916557591,
|
|
"loss": 6.1211,
|
|
"mean_token_accuracy": 0.1399856761097908,
|
|
"num_tokens": 3068996.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.157318639755249,
|
|
"epoch": 1.3534756928668787,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000497350184142062,
|
|
"loss": 5.9307,
|
|
"mean_token_accuracy": 0.15413330644369125,
|
|
"num_tokens": 3078991.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.14020094871521,
|
|
"epoch": 1.3580190822353475,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004972958280013886,
|
|
"loss": 5.8893,
|
|
"mean_token_accuracy": 0.14595961272716523,
|
|
"num_tokens": 3090200.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.1220416069030765,
|
|
"epoch": 1.3625624716038165,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004972409233678575,
|
|
"loss": 5.9451,
|
|
"mean_token_accuracy": 0.1446746289730072,
|
|
"num_tokens": 3101971.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.3625624716038165,
|
|
"eval_entropy": 5.998006772994995,
|
|
"eval_loss": 6.173778057098389,
|
|
"eval_mean_token_accuracy": 0.1402728841141906,
|
|
"eval_num_tokens": 3101971.0,
|
|
"eval_runtime": 2.0008,
|
|
"eval_samples_per_second": 1716.839,
|
|
"eval_steps_per_second": 214.917,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.163672637939453,
|
|
"epoch": 1.3671058609722853,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004971854703769401,
|
|
"loss": 6.0209,
|
|
"mean_token_accuracy": 0.1420085422694683,
|
|
"num_tokens": 3113381.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.293332433700561,
|
|
"epoch": 1.3716492503407542,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004971294691654616,
|
|
"loss": 5.9969,
|
|
"mean_token_accuracy": 0.1459997057914734,
|
|
"num_tokens": 3125099.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.071393442153931,
|
|
"epoch": 1.376192639709223,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000497072919871599,
|
|
"loss": 5.8885,
|
|
"mean_token_accuracy": 0.14934994727373124,
|
|
"num_tokens": 3135007.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.179635763168335,
|
|
"epoch": 1.380736029077692,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004970158226348824,
|
|
"loss": 5.9371,
|
|
"mean_token_accuracy": 0.14661285355687143,
|
|
"num_tokens": 3145031.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.185154438018799,
|
|
"epoch": 1.3852794184461608,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004969581775961932,
|
|
"loss": 5.9785,
|
|
"mean_token_accuracy": 0.14674355909228326,
|
|
"num_tokens": 3156044.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.160639667510987,
|
|
"epoch": 1.3898228078146297,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004968999848977651,
|
|
"loss": 5.8743,
|
|
"mean_token_accuracy": 0.15042117089033127,
|
|
"num_tokens": 3165124.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.065199089050293,
|
|
"epoch": 1.3943661971830985,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004968412446831827,
|
|
"loss": 5.8192,
|
|
"mean_token_accuracy": 0.15471493005752562,
|
|
"num_tokens": 3175246.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.1831460952758786,
|
|
"epoch": 1.3989095865515675,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004967819570973816,
|
|
"loss": 5.9167,
|
|
"mean_token_accuracy": 0.15386194437742234,
|
|
"num_tokens": 3184817.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.0874436378479,
|
|
"epoch": 1.4034529759200365,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000496722122286648,
|
|
"loss": 5.9543,
|
|
"mean_token_accuracy": 0.13574737310409546,
|
|
"num_tokens": 3195142.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.136379337310791,
|
|
"epoch": 1.4079963652885052,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004966617403986185,
|
|
"loss": 5.8976,
|
|
"mean_token_accuracy": 0.14391313940286637,
|
|
"num_tokens": 3205724.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.181877279281617,
|
|
"epoch": 1.412539754656974,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004966008115822793,
|
|
"loss": 6.0459,
|
|
"mean_token_accuracy": 0.1418856129050255,
|
|
"num_tokens": 3216919.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.292528104782105,
|
|
"epoch": 1.417083144025443,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004965393359879662,
|
|
"loss": 6.0416,
|
|
"mean_token_accuracy": 0.14282304048538208,
|
|
"num_tokens": 3228929.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.154995155334473,
|
|
"epoch": 1.421626533393912,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004964773137673641,
|
|
"loss": 5.9996,
|
|
"mean_token_accuracy": 0.14393652155995368,
|
|
"num_tokens": 3239541.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.1706877708435055,
|
|
"epoch": 1.4261699227623807,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004964147450735067,
|
|
"loss": 5.8848,
|
|
"mean_token_accuracy": 0.14642009437084197,
|
|
"num_tokens": 3249527.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.04834885597229,
|
|
"epoch": 1.4307133121308495,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000496351630060776,
|
|
"loss": 5.7945,
|
|
"mean_token_accuracy": 0.14744008034467698,
|
|
"num_tokens": 3260497.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.257424974441529,
|
|
"epoch": 1.4352567014993185,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004962879688849021,
|
|
"loss": 5.9799,
|
|
"mean_token_accuracy": 0.14823087900877,
|
|
"num_tokens": 3270865.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.060004806518554,
|
|
"epoch": 1.4398000908677875,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004962237617029625,
|
|
"loss": 5.9086,
|
|
"mean_token_accuracy": 0.1426824726164341,
|
|
"num_tokens": 3281531.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.207708930969238,
|
|
"epoch": 1.4443434802362562,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004961590086733821,
|
|
"loss": 5.8779,
|
|
"mean_token_accuracy": 0.14353057891130447,
|
|
"num_tokens": 3291970.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.059498691558838,
|
|
"epoch": 1.448886869604725,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004960937099559326,
|
|
"loss": 5.9209,
|
|
"mean_token_accuracy": 0.15407329946756362,
|
|
"num_tokens": 3301638.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.249056720733643,
|
|
"epoch": 1.453430258973194,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004960278657117321,
|
|
"loss": 6.0004,
|
|
"mean_token_accuracy": 0.1454174779355526,
|
|
"num_tokens": 3311780.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.074241209030151,
|
|
"epoch": 1.457973648341663,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004959614761032446,
|
|
"loss": 5.8559,
|
|
"mean_token_accuracy": 0.14448125958442687,
|
|
"num_tokens": 3322096.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.243053770065307,
|
|
"epoch": 1.4625170377101318,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00049589454129428,
|
|
"loss": 6.0287,
|
|
"mean_token_accuracy": 0.14130257591605186,
|
|
"num_tokens": 3332324.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.052849674224854,
|
|
"epoch": 1.4670604270786005,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004958270614499932,
|
|
"loss": 5.8215,
|
|
"mean_token_accuracy": 0.15624394565820693,
|
|
"num_tokens": 3342037.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.178307437896729,
|
|
"epoch": 1.4716038164470695,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004957590367368841,
|
|
"loss": 5.9252,
|
|
"mean_token_accuracy": 0.1461096227169037,
|
|
"num_tokens": 3353047.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.08671293258667,
|
|
"epoch": 1.4761472058155385,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004956904673227968,
|
|
"loss": 5.8278,
|
|
"mean_token_accuracy": 0.14527380913496019,
|
|
"num_tokens": 3363841.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.1023805141448975,
|
|
"epoch": 1.4806905951840073,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004956213533769195,
|
|
"loss": 5.9244,
|
|
"mean_token_accuracy": 0.14610948264598847,
|
|
"num_tokens": 3374086.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.123172569274902,
|
|
"epoch": 1.485233984552476,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004955516950697843,
|
|
"loss": 5.8951,
|
|
"mean_token_accuracy": 0.14830472320318222,
|
|
"num_tokens": 3384303.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.119581699371338,
|
|
"epoch": 1.489777373920945,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004954814925732657,
|
|
"loss": 5.8746,
|
|
"mean_token_accuracy": 0.14966145604848863,
|
|
"num_tokens": 3394858.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.098552131652832,
|
|
"epoch": 1.494320763289414,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004954107460605816,
|
|
"loss": 5.8808,
|
|
"mean_token_accuracy": 0.153575536608696,
|
|
"num_tokens": 3405252.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.24327540397644,
|
|
"epoch": 1.4988641526578828,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004953394557062922,
|
|
"loss": 5.9472,
|
|
"mean_token_accuracy": 0.14877036213874817,
|
|
"num_tokens": 3416007.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.228945350646972,
|
|
"epoch": 1.5034075420263515,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004952676216862989,
|
|
"loss": 5.9792,
|
|
"mean_token_accuracy": 0.14415789693593978,
|
|
"num_tokens": 3426932.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.038790559768676,
|
|
"epoch": 1.5079509313948205,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004951952441778453,
|
|
"loss": 5.8586,
|
|
"mean_token_accuracy": 0.15109123438596725,
|
|
"num_tokens": 3437262.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.09799542427063,
|
|
"epoch": 1.5124943207632895,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004951223233595158,
|
|
"loss": 5.8273,
|
|
"mean_token_accuracy": 0.15195785015821456,
|
|
"num_tokens": 3447525.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.1701806545257565,
|
|
"epoch": 1.5170377101317583,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000495048859411235,
|
|
"loss": 5.9336,
|
|
"mean_token_accuracy": 0.15256594866514206,
|
|
"num_tokens": 3458198.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.0774983882904055,
|
|
"epoch": 1.521581099500227,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004949748525142681,
|
|
"loss": 5.8165,
|
|
"mean_token_accuracy": 0.1567363977432251,
|
|
"num_tokens": 3468149.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.081609010696411,
|
|
"epoch": 1.526124488868696,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004949003028512195,
|
|
"loss": 5.8139,
|
|
"mean_token_accuracy": 0.14687097519636155,
|
|
"num_tokens": 3478355.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.097584533691406,
|
|
"epoch": 1.530667878237165,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004948252106060334,
|
|
"loss": 5.8428,
|
|
"mean_token_accuracy": 0.15158360898494722,
|
|
"num_tokens": 3488726.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.048998165130615,
|
|
"epoch": 1.5352112676056338,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004947495759639923,
|
|
"loss": 5.7861,
|
|
"mean_token_accuracy": 0.15492362827062606,
|
|
"num_tokens": 3498346.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.995868158340454,
|
|
"epoch": 1.5397546569741025,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004946733991117171,
|
|
"loss": 5.8739,
|
|
"mean_token_accuracy": 0.15013641715049744,
|
|
"num_tokens": 3508213.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.236345434188843,
|
|
"epoch": 1.5442980463425715,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004945966802371668,
|
|
"loss": 5.9989,
|
|
"mean_token_accuracy": 0.15021879523992537,
|
|
"num_tokens": 3518978.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.119385671615601,
|
|
"epoch": 1.5488414357110405,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004945194195296374,
|
|
"loss": 5.9327,
|
|
"mean_token_accuracy": 0.1489047683775425,
|
|
"num_tokens": 3529328.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.03963418006897,
|
|
"epoch": 1.5533848250795095,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000494441617179762,
|
|
"loss": 5.8623,
|
|
"mean_token_accuracy": 0.14225288853049278,
|
|
"num_tokens": 3539347.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.159651279449463,
|
|
"epoch": 1.5579282144479782,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004943632733795104,
|
|
"loss": 5.8838,
|
|
"mean_token_accuracy": 0.1437460221350193,
|
|
"num_tokens": 3549798.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.087096071243286,
|
|
"epoch": 1.562471603816447,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004942843883221879,
|
|
"loss": 5.978,
|
|
"mean_token_accuracy": 0.1412530466914177,
|
|
"num_tokens": 3560363.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.1436097621917725,
|
|
"epoch": 1.567014993184916,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004942049622024357,
|
|
"loss": 5.9047,
|
|
"mean_token_accuracy": 0.1528037503361702,
|
|
"num_tokens": 3570080.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.058519458770752,
|
|
"epoch": 1.571558382553385,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004941249952162298,
|
|
"loss": 5.9049,
|
|
"mean_token_accuracy": 0.14703553318977355,
|
|
"num_tokens": 3580053.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.13126769065857,
|
|
"epoch": 1.5761017719218537,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004940444875608807,
|
|
"loss": 5.9875,
|
|
"mean_token_accuracy": 0.14530991315841674,
|
|
"num_tokens": 3590119.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.181125068664551,
|
|
"epoch": 1.5806451612903225,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004939634394350332,
|
|
"loss": 5.9674,
|
|
"mean_token_accuracy": 0.1385622426867485,
|
|
"num_tokens": 3601809.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.158671808242798,
|
|
"epoch": 1.5851885506587915,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004938818510386654,
|
|
"loss": 5.8219,
|
|
"mean_token_accuracy": 0.15508615523576735,
|
|
"num_tokens": 3611752.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.0497949600219725,
|
|
"epoch": 1.5897319400272605,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004937997225730886,
|
|
"loss": 5.8889,
|
|
"mean_token_accuracy": 0.14583897739648818,
|
|
"num_tokens": 3621137.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.117823123931885,
|
|
"epoch": 1.5942753293957292,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004937170542409464,
|
|
"loss": 5.8594,
|
|
"mean_token_accuracy": 0.15217961221933365,
|
|
"num_tokens": 3631567.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.027679061889648,
|
|
"epoch": 1.598818718764198,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000493633846246215,
|
|
"loss": 5.8221,
|
|
"mean_token_accuracy": 0.14410740211606027,
|
|
"num_tokens": 3641145.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.09434552192688,
|
|
"epoch": 1.603362108132667,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004935500987942018,
|
|
"loss": 5.8681,
|
|
"mean_token_accuracy": 0.1448398560285568,
|
|
"num_tokens": 3651716.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.130527305603027,
|
|
"epoch": 1.607905497501136,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004934658120915452,
|
|
"loss": 5.8207,
|
|
"mean_token_accuracy": 0.15499569922685624,
|
|
"num_tokens": 3662577.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.012383270263672,
|
|
"epoch": 1.6124488868696047,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004933809863462142,
|
|
"loss": 5.8048,
|
|
"mean_token_accuracy": 0.15538352131843566,
|
|
"num_tokens": 3672984.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.1079894542694095,
|
|
"epoch": 1.6169922762380735,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004932956217675082,
|
|
"loss": 5.8837,
|
|
"mean_token_accuracy": 0.15415330082178116,
|
|
"num_tokens": 3682634.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.053774452209472,
|
|
"epoch": 1.6215356656065425,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004932097185660555,
|
|
"loss": 5.8568,
|
|
"mean_token_accuracy": 0.14938003569841385,
|
|
"num_tokens": 3693022.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.923639822006225,
|
|
"epoch": 1.6260790549750115,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004931232769538138,
|
|
"loss": 5.7328,
|
|
"mean_token_accuracy": 0.15786038786172868,
|
|
"num_tokens": 3702234.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.0286894798278805,
|
|
"epoch": 1.6306224443434802,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004930362971440692,
|
|
"loss": 5.9116,
|
|
"mean_token_accuracy": 0.1445778265595436,
|
|
"num_tokens": 3712235.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.143106698989868,
|
|
"epoch": 1.635165833711949,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004929487793514358,
|
|
"loss": 5.8417,
|
|
"mean_token_accuracy": 0.15846360027790068,
|
|
"num_tokens": 3722745.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.9879150390625,
|
|
"epoch": 1.639709223080418,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000492860723791855,
|
|
"loss": 5.8976,
|
|
"mean_token_accuracy": 0.15017399489879607,
|
|
"num_tokens": 3732942.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.137276124954224,
|
|
"epoch": 1.644252612448887,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004927721306825952,
|
|
"loss": 5.8819,
|
|
"mean_token_accuracy": 0.1497318923473358,
|
|
"num_tokens": 3743806.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.006168842315674,
|
|
"epoch": 1.6487960018173557,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004926830002422509,
|
|
"loss": 5.8719,
|
|
"mean_token_accuracy": 0.15341466218233107,
|
|
"num_tokens": 3755115.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.078045082092285,
|
|
"epoch": 1.6533393911858245,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000492593332690743,
|
|
"loss": 5.8205,
|
|
"mean_token_accuracy": 0.15602415949106216,
|
|
"num_tokens": 3765867.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.980977249145508,
|
|
"epoch": 1.6578827805542935,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004925031282493167,
|
|
"loss": 5.8035,
|
|
"mean_token_accuracy": 0.15344792902469634,
|
|
"num_tokens": 3777071.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.016601514816284,
|
|
"epoch": 1.6624261699227625,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004924123871405428,
|
|
"loss": 5.7974,
|
|
"mean_token_accuracy": 0.1537347137928009,
|
|
"num_tokens": 3786554.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.102873992919922,
|
|
"epoch": 1.6669695592912313,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004923211095883161,
|
|
"loss": 5.8517,
|
|
"mean_token_accuracy": 0.15107907652854918,
|
|
"num_tokens": 3797627.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.971213245391846,
|
|
"epoch": 1.6715129486597,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004922292958178546,
|
|
"loss": 5.8426,
|
|
"mean_token_accuracy": 0.1507897347211838,
|
|
"num_tokens": 3808841.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.1389758586883545,
|
|
"epoch": 1.676056338028169,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004921369460557,
|
|
"loss": 5.8725,
|
|
"mean_token_accuracy": 0.14769963175058365,
|
|
"num_tokens": 3818995.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.062976360321045,
|
|
"epoch": 1.680599727396638,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004920440605297159,
|
|
"loss": 5.9237,
|
|
"mean_token_accuracy": 0.1460937276482582,
|
|
"num_tokens": 3828762.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.094574308395385,
|
|
"epoch": 1.6851431167651068,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000491950639469088,
|
|
"loss": 5.7671,
|
|
"mean_token_accuracy": 0.15168659314513205,
|
|
"num_tokens": 3838752.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.985349941253662,
|
|
"epoch": 1.6896865061335755,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004918566831043238,
|
|
"loss": 5.8084,
|
|
"mean_token_accuracy": 0.15743009150028228,
|
|
"num_tokens": 3849664.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.03216323852539,
|
|
"epoch": 1.6942298955020445,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000491762191667251,
|
|
"loss": 5.8003,
|
|
"mean_token_accuracy": 0.1560536116361618,
|
|
"num_tokens": 3859053.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.0026287078857425,
|
|
"epoch": 1.6987732848705135,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000491667165391018,
|
|
"loss": 5.8584,
|
|
"mean_token_accuracy": 0.14554656594991683,
|
|
"num_tokens": 3869654.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.073776149749756,
|
|
"epoch": 1.7033166742389823,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004915716045100927,
|
|
"loss": 5.8558,
|
|
"mean_token_accuracy": 0.14700101166963578,
|
|
"num_tokens": 3879678.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.020952033996582,
|
|
"epoch": 1.707860063607451,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004914755092602621,
|
|
"loss": 5.7539,
|
|
"mean_token_accuracy": 0.15705975145101547,
|
|
"num_tokens": 3889950.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.9746153354644775,
|
|
"epoch": 1.71240345297592,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004913788798786314,
|
|
"loss": 5.838,
|
|
"mean_token_accuracy": 0.15099019408226014,
|
|
"num_tokens": 3900095.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.169710397720337,
|
|
"epoch": 1.716946842344389,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004912817166036243,
|
|
"loss": 5.8931,
|
|
"mean_token_accuracy": 0.14681051373481752,
|
|
"num_tokens": 3911034.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.995103216171264,
|
|
"epoch": 1.7214902317128578,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004911840196749813,
|
|
"loss": 5.8633,
|
|
"mean_token_accuracy": 0.15044671893119813,
|
|
"num_tokens": 3922219.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.01338906288147,
|
|
"epoch": 1.7260336210813265,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004910857893337602,
|
|
"loss": 5.8327,
|
|
"mean_token_accuracy": 0.15175913274288177,
|
|
"num_tokens": 3932235.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.006257343292236,
|
|
"epoch": 1.7305770104497955,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004909870258223342,
|
|
"loss": 5.7738,
|
|
"mean_token_accuracy": 0.1562149293720722,
|
|
"num_tokens": 3941845.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.0315406799316404,
|
|
"epoch": 1.7351203998182645,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004908877293843926,
|
|
"loss": 5.8632,
|
|
"mean_token_accuracy": 0.14723684638738632,
|
|
"num_tokens": 3953392.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.0056493282318115,
|
|
"epoch": 1.7396637891867333,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004907879002649395,
|
|
"loss": 5.7197,
|
|
"mean_token_accuracy": 0.15770288556814194,
|
|
"num_tokens": 3963294.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.951028156280517,
|
|
"epoch": 1.744207178555202,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004906875387102935,
|
|
"loss": 5.7917,
|
|
"mean_token_accuracy": 0.1482487738132477,
|
|
"num_tokens": 3973660.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.048778867721557,
|
|
"epoch": 1.748750567923671,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004905866449680864,
|
|
"loss": 5.7858,
|
|
"mean_token_accuracy": 0.15318481773138046,
|
|
"num_tokens": 3983587.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.014044857025146,
|
|
"epoch": 1.75329395729214,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004904852192872638,
|
|
"loss": 5.8483,
|
|
"mean_token_accuracy": 0.1447431579232216,
|
|
"num_tokens": 3994346.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.030064249038697,
|
|
"epoch": 1.757837346660609,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004903832619180834,
|
|
"loss": 5.8229,
|
|
"mean_token_accuracy": 0.15561417788267135,
|
|
"num_tokens": 4005234.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.914287424087524,
|
|
"epoch": 1.7623807360290777,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004902807731121147,
|
|
"loss": 5.7382,
|
|
"mean_token_accuracy": 0.15640076696872712,
|
|
"num_tokens": 4015542.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.9699568271636965,
|
|
"epoch": 1.7669241253975465,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004901777531222389,
|
|
"loss": 5.7553,
|
|
"mean_token_accuracy": 0.16071913093328477,
|
|
"num_tokens": 4024874.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.061671257019043,
|
|
"epoch": 1.7714675147660155,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004900742022026474,
|
|
"loss": 5.9002,
|
|
"mean_token_accuracy": 0.15215046256780623,
|
|
"num_tokens": 4035880.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.04217495918274,
|
|
"epoch": 1.7760109041344845,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004899701206088419,
|
|
"loss": 5.7851,
|
|
"mean_token_accuracy": 0.15591228753328323,
|
|
"num_tokens": 4046484.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.0316204071044925,
|
|
"epoch": 1.7805542935029532,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004898655085976333,
|
|
"loss": 5.8441,
|
|
"mean_token_accuracy": 0.15250226184725763,
|
|
"num_tokens": 4057037.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.102791118621826,
|
|
"epoch": 1.785097682871422,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004897603664271413,
|
|
"loss": 5.9278,
|
|
"mean_token_accuracy": 0.1487814873456955,
|
|
"num_tokens": 4066169.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.98809928894043,
|
|
"epoch": 1.789641072239891,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000489654694356794,
|
|
"loss": 5.8312,
|
|
"mean_token_accuracy": 0.14956735149025918,
|
|
"num_tokens": 4076697.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.998944187164307,
|
|
"epoch": 1.79418446160836,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004895484926473265,
|
|
"loss": 5.7309,
|
|
"mean_token_accuracy": 0.15552960187196732,
|
|
"num_tokens": 4087372.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.047774410247802,
|
|
"epoch": 1.7987278509768287,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000489441761560781,
|
|
"loss": 5.8148,
|
|
"mean_token_accuracy": 0.14908257573843003,
|
|
"num_tokens": 4097653.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.006575441360473,
|
|
"epoch": 1.8032712403452975,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000489334501360506,
|
|
"loss": 5.7478,
|
|
"mean_token_accuracy": 0.15341038405895233,
|
|
"num_tokens": 4107381.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.064118814468384,
|
|
"epoch": 1.8078146297137665,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004892267123111551,
|
|
"loss": 5.8693,
|
|
"mean_token_accuracy": 0.1496926262974739,
|
|
"num_tokens": 4118762.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.946984529495239,
|
|
"epoch": 1.8123580190822355,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004891183946786873,
|
|
"loss": 5.7685,
|
|
"mean_token_accuracy": 0.15168437510728836,
|
|
"num_tokens": 4129011.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.085385131835937,
|
|
"epoch": 1.8169014084507042,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004890095487303655,
|
|
"loss": 5.8416,
|
|
"mean_token_accuracy": 0.14939511120319365,
|
|
"num_tokens": 4138975.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.8169014084507042,
|
|
"eval_entropy": 5.802581080725027,
|
|
"eval_loss": 5.989554405212402,
|
|
"eval_mean_token_accuracy": 0.15064924421698547,
|
|
"eval_num_tokens": 4138975.0,
|
|
"eval_runtime": 2.0074,
|
|
"eval_samples_per_second": 1711.137,
|
|
"eval_steps_per_second": 214.203,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.994388818740845,
|
|
"epoch": 1.821444797819173,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004889001747347562,
|
|
"loss": 5.8601,
|
|
"mean_token_accuracy": 0.14598984345793725,
|
|
"num_tokens": 4150164.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.953599166870117,
|
|
"epoch": 1.825988187187642,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000488790272961729,
|
|
"loss": 5.762,
|
|
"mean_token_accuracy": 0.15967332124710082,
|
|
"num_tokens": 4159872.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.017344665527344,
|
|
"epoch": 1.830531576556111,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004886798436824556,
|
|
"loss": 5.8409,
|
|
"mean_token_accuracy": 0.14872767329216002,
|
|
"num_tokens": 4170665.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.01618185043335,
|
|
"epoch": 1.8350749659245797,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004885688871694091,
|
|
"loss": 5.7142,
|
|
"mean_token_accuracy": 0.1626022905111313,
|
|
"num_tokens": 4180476.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.995317649841309,
|
|
"epoch": 1.8396183552930485,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004884574036963639,
|
|
"loss": 5.8566,
|
|
"mean_token_accuracy": 0.15099751204252243,
|
|
"num_tokens": 4190740.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.956916618347168,
|
|
"epoch": 1.8441617446615175,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004883453935383941,
|
|
"loss": 5.7434,
|
|
"mean_token_accuracy": 0.1571941539645195,
|
|
"num_tokens": 4201504.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.875329256057739,
|
|
"epoch": 1.8487051340299865,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004882328569718738,
|
|
"loss": 5.6753,
|
|
"mean_token_accuracy": 0.16390404999256133,
|
|
"num_tokens": 4211614.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.969325065612793,
|
|
"epoch": 1.8532485233984552,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004881197942744758,
|
|
"loss": 5.8008,
|
|
"mean_token_accuracy": 0.15329188108444214,
|
|
"num_tokens": 4221141.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.991888856887817,
|
|
"epoch": 1.857791912766924,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004880062057251711,
|
|
"loss": 5.8352,
|
|
"mean_token_accuracy": 0.15410455614328383,
|
|
"num_tokens": 4231877.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.914742946624756,
|
|
"epoch": 1.862335302135393,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000487892091604228,
|
|
"loss": 5.6553,
|
|
"mean_token_accuracy": 0.16117654740810394,
|
|
"num_tokens": 4242331.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.914829111099243,
|
|
"epoch": 1.866878691503862,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048777745219321186,
|
|
"loss": 5.6845,
|
|
"mean_token_accuracy": 0.1546552672982216,
|
|
"num_tokens": 4252406.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.974059200286865,
|
|
"epoch": 1.8714220808723308,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048766228777498405,
|
|
"loss": 5.7597,
|
|
"mean_token_accuracy": 0.1531647711992264,
|
|
"num_tokens": 4263036.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.9508284568786625,
|
|
"epoch": 1.8759654702407995,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048754659863370125,
|
|
"loss": 5.7335,
|
|
"mean_token_accuracy": 0.1608620196580887,
|
|
"num_tokens": 4273032.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.926699542999268,
|
|
"epoch": 1.8805088596092685,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048743038505481503,
|
|
"loss": 5.8057,
|
|
"mean_token_accuracy": 0.1554887369275093,
|
|
"num_tokens": 4283360.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.959084749221802,
|
|
"epoch": 1.8850522489777375,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004873136473250708,
|
|
"loss": 5.7423,
|
|
"mean_token_accuracy": 0.15936065316200257,
|
|
"num_tokens": 4292623.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.969320487976074,
|
|
"epoch": 1.8895956383462063,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048719638573250715,
|
|
"loss": 5.7658,
|
|
"mean_token_accuracy": 0.1576855793595314,
|
|
"num_tokens": 4302817.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.005345106124878,
|
|
"epoch": 1.894139027714675,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004870786005664555,
|
|
"loss": 5.7671,
|
|
"mean_token_accuracy": 0.15660459101200103,
|
|
"num_tokens": 4313049.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.913744068145752,
|
|
"epoch": 1.898682417083144,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004869602921175392,
|
|
"loss": 5.6484,
|
|
"mean_token_accuracy": 0.1619171142578125,
|
|
"num_tokens": 4322873.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.8904248714447025,
|
|
"epoch": 1.903225806451613,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004868414606776724,
|
|
"loss": 5.8048,
|
|
"mean_token_accuracy": 0.14883103221654892,
|
|
"num_tokens": 4333339.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.974893951416016,
|
|
"epoch": 1.9077691958200818,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00048672210654005996,
|
|
"loss": 5.7668,
|
|
"mean_token_accuracy": 0.15438382625579833,
|
|
"num_tokens": 4344917.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.866996431350708,
|
|
"epoch": 1.9123125851885505,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048660222999919616,
|
|
"loss": 5.6211,
|
|
"mean_token_accuracy": 0.16387978196144104,
|
|
"num_tokens": 4354140.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.947313690185547,
|
|
"epoch": 1.9168559745570195,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004864818313508647,
|
|
"loss": 5.871,
|
|
"mean_token_accuracy": 0.14966592639684678,
|
|
"num_tokens": 4364976.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.965764045715332,
|
|
"epoch": 1.9213993639254885,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004863609108921372,
|
|
"loss": 5.6455,
|
|
"mean_token_accuracy": 0.16220821142196656,
|
|
"num_tokens": 4376423.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.96432089805603,
|
|
"epoch": 1.9259427532939573,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004862394689213728,
|
|
"loss": 5.7859,
|
|
"mean_token_accuracy": 0.15559903681278228,
|
|
"num_tokens": 4387865.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.928690385818482,
|
|
"epoch": 1.930486142662426,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004861175057382175,
|
|
"loss": 5.7782,
|
|
"mean_token_accuracy": 0.1554933860898018,
|
|
"num_tokens": 4398275.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.014926385879517,
|
|
"epoch": 1.935029532030895,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004859950216436034,
|
|
"loss": 5.705,
|
|
"mean_token_accuracy": 0.15233548432588578,
|
|
"num_tokens": 4407396.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.056639719009399,
|
|
"epoch": 1.939572921399364,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004858720169397478,
|
|
"loss": 5.9496,
|
|
"mean_token_accuracy": 0.147593604773283,
|
|
"num_tokens": 4419682.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.961567783355713,
|
|
"epoch": 1.9441163107678328,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004857484919301525,
|
|
"loss": 5.8116,
|
|
"mean_token_accuracy": 0.1537695899605751,
|
|
"num_tokens": 4430590.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.9822437286376955,
|
|
"epoch": 1.9486597001363015,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004856244469196034,
|
|
"loss": 5.7355,
|
|
"mean_token_accuracy": 0.158633291721344,
|
|
"num_tokens": 4441979.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.946609210968018,
|
|
"epoch": 1.9532030895047705,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000485499882214169,
|
|
"loss": 5.7522,
|
|
"mean_token_accuracy": 0.16035017371177673,
|
|
"num_tokens": 4451975.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.85480809211731,
|
|
"epoch": 1.9577464788732395,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004853747981212003,
|
|
"loss": 5.6904,
|
|
"mean_token_accuracy": 0.1584114506840706,
|
|
"num_tokens": 4462778.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.0221734046936035,
|
|
"epoch": 1.9622898682417085,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004852491949493301,
|
|
"loss": 5.7479,
|
|
"mean_token_accuracy": 0.16584970504045488,
|
|
"num_tokens": 4472461.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.8947083950042725,
|
|
"epoch": 1.9668332576101772,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004851230730084716,
|
|
"loss": 5.6977,
|
|
"mean_token_accuracy": 0.15985091775655746,
|
|
"num_tokens": 4482507.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.9333573341369625,
|
|
"epoch": 1.971376646978646,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048499643260981816,
|
|
"loss": 5.72,
|
|
"mean_token_accuracy": 0.15951878279447557,
|
|
"num_tokens": 4492601.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.92025089263916,
|
|
"epoch": 1.975920036347115,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004848692740658424,
|
|
"loss": 5.8213,
|
|
"mean_token_accuracy": 0.15883293896913528,
|
|
"num_tokens": 4503696.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.95161771774292,
|
|
"epoch": 1.980463425715584,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004847415976902954,
|
|
"loss": 5.7105,
|
|
"mean_token_accuracy": 0.15256518721580506,
|
|
"num_tokens": 4513649.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.954947519302368,
|
|
"epoch": 1.9850068150840527,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048461340379820587,
|
|
"loss": 5.733,
|
|
"mean_token_accuracy": 0.15729330182075502,
|
|
"num_tokens": 4524054.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.966910886764526,
|
|
"epoch": 1.9895502044525215,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048448469270587945,
|
|
"loss": 5.7227,
|
|
"mean_token_accuracy": 0.15084918439388276,
|
|
"num_tokens": 4535549.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.9350450992584225,
|
|
"epoch": 1.9940935938209905,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048435546473089814,
|
|
"loss": 5.7532,
|
|
"mean_token_accuracy": 0.15482822358608245,
|
|
"num_tokens": 4545654.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.928518342971802,
|
|
"epoch": 1.9986369831894595,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048422572019211896,
|
|
"loss": 5.757,
|
|
"mean_token_accuracy": 0.15873578190803528,
|
|
"num_tokens": 4556230.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.900713867611355,
|
|
"epoch": 2.0027260336210815,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00048409545940967375,
|
|
"loss": 5.5157,
|
|
"mean_token_accuracy": 0.16549871199660832,
|
|
"num_tokens": 4564486.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.888030290603638,
|
|
"epoch": 2.00726942298955,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000483964682704968,
|
|
"loss": 5.4203,
|
|
"mean_token_accuracy": 0.171760892868042,
|
|
"num_tokens": 4575026.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.825547981262207,
|
|
"epoch": 2.011812812358019,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048383339040068024,
|
|
"loss": 5.4509,
|
|
"mean_token_accuracy": 0.16520898640155793,
|
|
"num_tokens": 4585902.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.032274007797241,
|
|
"epoch": 2.016356201726488,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004837015828207612,
|
|
"loss": 5.5558,
|
|
"mean_token_accuracy": 0.1652931973338127,
|
|
"num_tokens": 4597609.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.861545181274414,
|
|
"epoch": 2.020899591094957,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004835692602904329,
|
|
"loss": 5.4229,
|
|
"mean_token_accuracy": 0.17381911724805832,
|
|
"num_tokens": 4607531.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.8608067512512205,
|
|
"epoch": 2.0254429804634255,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004834364231361882,
|
|
"loss": 5.4707,
|
|
"mean_token_accuracy": 0.16835336834192277,
|
|
"num_tokens": 4618826.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.993276453018188,
|
|
"epoch": 2.0299863698318945,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004833030716857894,
|
|
"loss": 5.4663,
|
|
"mean_token_accuracy": 0.17149772942066194,
|
|
"num_tokens": 4629526.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.863670110702515,
|
|
"epoch": 2.0345297592003635,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048316920626826807,
|
|
"loss": 5.499,
|
|
"mean_token_accuracy": 0.17145546674728393,
|
|
"num_tokens": 4639930.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.893801164627075,
|
|
"epoch": 2.0390731485688325,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048303482721392363,
|
|
"loss": 5.398,
|
|
"mean_token_accuracy": 0.17716680318117142,
|
|
"num_tokens": 4649661.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.778148555755616,
|
|
"epoch": 2.043616537937301,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00048289993485432325,
|
|
"loss": 5.3957,
|
|
"mean_token_accuracy": 0.17859122008085251,
|
|
"num_tokens": 4660178.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.8800742626190186,
|
|
"epoch": 2.04815992730577,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048276452952230036,
|
|
"loss": 5.4478,
|
|
"mean_token_accuracy": 0.17156661301851273,
|
|
"num_tokens": 4670417.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.827249670028687,
|
|
"epoch": 2.052703316674239,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048262861155195407,
|
|
"loss": 5.3955,
|
|
"mean_token_accuracy": 0.18324804604053496,
|
|
"num_tokens": 4680221.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.834317541122436,
|
|
"epoch": 2.057246706042708,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004824921812786486,
|
|
"loss": 5.4471,
|
|
"mean_token_accuracy": 0.16734634935855866,
|
|
"num_tokens": 4691136.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.754065322875976,
|
|
"epoch": 2.0617900954111765,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000482355239039012,
|
|
"loss": 5.2977,
|
|
"mean_token_accuracy": 0.18202615827322005,
|
|
"num_tokens": 4700879.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.814774465560913,
|
|
"epoch": 2.0663334847796455,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004822177851709356,
|
|
"loss": 5.4785,
|
|
"mean_token_accuracy": 0.16296902894973755,
|
|
"num_tokens": 4711477.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.825762462615967,
|
|
"epoch": 2.0708768741481145,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048207982001357327,
|
|
"loss": 5.4469,
|
|
"mean_token_accuracy": 0.17125846296548844,
|
|
"num_tokens": 4721973.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.7884949207305905,
|
|
"epoch": 2.0754202635165835,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004819413439073402,
|
|
"loss": 5.4509,
|
|
"mean_token_accuracy": 0.1706570252776146,
|
|
"num_tokens": 4732425.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.817496299743652,
|
|
"epoch": 2.079963652885052,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004818023571939125,
|
|
"loss": 5.4223,
|
|
"mean_token_accuracy": 0.1764614388346672,
|
|
"num_tokens": 4743377.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.892448902130127,
|
|
"epoch": 2.084507042253521,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000481662860216226,
|
|
"loss": 5.4782,
|
|
"mean_token_accuracy": 0.17242299020290375,
|
|
"num_tokens": 4753534.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.813091564178467,
|
|
"epoch": 2.08905043162199,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048152285331847574,
|
|
"loss": 5.425,
|
|
"mean_token_accuracy": 0.17044468671083451,
|
|
"num_tokens": 4763159.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.786837244033814,
|
|
"epoch": 2.093593820990459,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004813823368461148,
|
|
"loss": 5.3961,
|
|
"mean_token_accuracy": 0.17746685296297074,
|
|
"num_tokens": 4773497.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.8038170337677,
|
|
"epoch": 2.098137210358928,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004812413111458534,
|
|
"loss": 5.4938,
|
|
"mean_token_accuracy": 0.17549572438001632,
|
|
"num_tokens": 4784291.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.802530765533447,
|
|
"epoch": 2.1026805997273965,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004810997765656587,
|
|
"loss": 5.443,
|
|
"mean_token_accuracy": 0.17139201760292053,
|
|
"num_tokens": 4794762.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.9274169921875,
|
|
"epoch": 2.1072239890958655,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048095773345475294,
|
|
"loss": 5.5627,
|
|
"mean_token_accuracy": 0.16689665615558624,
|
|
"num_tokens": 4804449.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.812848997116089,
|
|
"epoch": 2.1117673784643345,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00048081518216361367,
|
|
"loss": 5.424,
|
|
"mean_token_accuracy": 0.17251452952623367,
|
|
"num_tokens": 4814172.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.852766227722168,
|
|
"epoch": 2.1163107678328035,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004806721230439719,
|
|
"loss": 5.5253,
|
|
"mean_token_accuracy": 0.17016936093568802,
|
|
"num_tokens": 4824909.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.9117193698883055,
|
|
"epoch": 2.120854157201272,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004805285564488117,
|
|
"loss": 5.5564,
|
|
"mean_token_accuracy": 0.17360106557607652,
|
|
"num_tokens": 4834546.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.7922993183135985,
|
|
"epoch": 2.125397546569741,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004803844827323696,
|
|
"loss": 5.4604,
|
|
"mean_token_accuracy": 0.16917911916971207,
|
|
"num_tokens": 4844809.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.79258623123169,
|
|
"epoch": 2.12994093593821,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004802399022501329,
|
|
"loss": 5.435,
|
|
"mean_token_accuracy": 0.1720902919769287,
|
|
"num_tokens": 4855883.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.873880672454834,
|
|
"epoch": 2.134484325306679,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004800948153588397,
|
|
"loss": 5.4778,
|
|
"mean_token_accuracy": 0.17590655088424684,
|
|
"num_tokens": 4865936.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.714124965667724,
|
|
"epoch": 2.1390277146751475,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047994922241647746,
|
|
"loss": 5.4071,
|
|
"mean_token_accuracy": 0.17288148552179336,
|
|
"num_tokens": 4876621.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.679599809646606,
|
|
"epoch": 2.1435711040436165,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047980312378228244,
|
|
"loss": 5.37,
|
|
"mean_token_accuracy": 0.16944972723722457,
|
|
"num_tokens": 4886315.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.838567876815796,
|
|
"epoch": 2.1481144934120855,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047965651981673823,
|
|
"loss": 5.4553,
|
|
"mean_token_accuracy": 0.1743234723806381,
|
|
"num_tokens": 4897214.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.813695907592773,
|
|
"epoch": 2.1526578827805545,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047950941088157576,
|
|
"loss": 5.4593,
|
|
"mean_token_accuracy": 0.1741379454731941,
|
|
"num_tokens": 4907360.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.7622902393341064,
|
|
"epoch": 2.157201272149023,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004793617973397715,
|
|
"loss": 5.42,
|
|
"mean_token_accuracy": 0.1762954756617546,
|
|
"num_tokens": 4917591.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.8187230110168455,
|
|
"epoch": 2.161744661517492,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047921367955554745,
|
|
"loss": 5.5005,
|
|
"mean_token_accuracy": 0.16116744875907899,
|
|
"num_tokens": 4928075.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.8125792980194095,
|
|
"epoch": 2.166288050885961,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004790650578943693,
|
|
"loss": 5.3972,
|
|
"mean_token_accuracy": 0.17632868140935898,
|
|
"num_tokens": 4937898.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.751837873458863,
|
|
"epoch": 2.17083144025443,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047891593272294636,
|
|
"loss": 5.4593,
|
|
"mean_token_accuracy": 0.16703200191259385,
|
|
"num_tokens": 4948152.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.857007837295532,
|
|
"epoch": 2.1753748296228985,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047876630440923,
|
|
"loss": 5.4436,
|
|
"mean_token_accuracy": 0.17320962250232697,
|
|
"num_tokens": 4957973.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.687740468978882,
|
|
"epoch": 2.1799182189913675,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004786161733224134,
|
|
"loss": 5.4024,
|
|
"mean_token_accuracy": 0.17448024451732635,
|
|
"num_tokens": 4967375.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.797864818572998,
|
|
"epoch": 2.1844616083598365,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047846553983293006,
|
|
"loss": 5.4203,
|
|
"mean_token_accuracy": 0.17751937955617905,
|
|
"num_tokens": 4977286.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.751899147033692,
|
|
"epoch": 2.1890049977283055,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047831440431245307,
|
|
"loss": 5.4798,
|
|
"mean_token_accuracy": 0.16854800581932067,
|
|
"num_tokens": 4987930.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.89511399269104,
|
|
"epoch": 2.193548387096774,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047816276713389424,
|
|
"loss": 5.5353,
|
|
"mean_token_accuracy": 0.16659025400876998,
|
|
"num_tokens": 4998707.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.739464187622071,
|
|
"epoch": 2.198091776465243,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004780106286714034,
|
|
"loss": 5.3842,
|
|
"mean_token_accuracy": 0.18243036270141602,
|
|
"num_tokens": 5008540.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.736773681640625,
|
|
"epoch": 2.202635165833712,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000477857989300367,
|
|
"loss": 5.4372,
|
|
"mean_token_accuracy": 0.17635186314582824,
|
|
"num_tokens": 5018199.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.849097681045532,
|
|
"epoch": 2.207178555202181,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047770484939740757,
|
|
"loss": 5.4636,
|
|
"mean_token_accuracy": 0.1730012848973274,
|
|
"num_tokens": 5028251.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.723977041244507,
|
|
"epoch": 2.2117219445706495,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004775512093403824,
|
|
"loss": 5.4572,
|
|
"mean_token_accuracy": 0.16673808097839354,
|
|
"num_tokens": 5038290.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.776793336868286,
|
|
"epoch": 2.2162653339391185,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00047739706950838327,
|
|
"loss": 5.4156,
|
|
"mean_token_accuracy": 0.1714222863316536,
|
|
"num_tokens": 5048337.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.822423934936523,
|
|
"epoch": 2.2208087233075875,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047724243028173484,
|
|
"loss": 5.4482,
|
|
"mean_token_accuracy": 0.16746363788843155,
|
|
"num_tokens": 5058533.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.7664214134216305,
|
|
"epoch": 2.2253521126760565,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047708729204199406,
|
|
"loss": 5.4585,
|
|
"mean_token_accuracy": 0.17420620322227479,
|
|
"num_tokens": 5068556.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.74825701713562,
|
|
"epoch": 2.229895502044525,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000476931655171949,
|
|
"loss": 5.5388,
|
|
"mean_token_accuracy": 0.1687641844153404,
|
|
"num_tokens": 5079317.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.872238731384277,
|
|
"epoch": 2.234438891412994,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004767755200556183,
|
|
"loss": 5.5164,
|
|
"mean_token_accuracy": 0.17064337730407714,
|
|
"num_tokens": 5089544.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.734923696517944,
|
|
"epoch": 2.238982280781463,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004766188870782497,
|
|
"loss": 5.503,
|
|
"mean_token_accuracy": 0.16758169382810592,
|
|
"num_tokens": 5099491.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.881393003463745,
|
|
"epoch": 2.243525670149932,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047646175662631967,
|
|
"loss": 5.5015,
|
|
"mean_token_accuracy": 0.16988374143838883,
|
|
"num_tokens": 5109523.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.796596050262451,
|
|
"epoch": 2.2480690595184005,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047630412908753183,
|
|
"loss": 5.4462,
|
|
"mean_token_accuracy": 0.17399886548519133,
|
|
"num_tokens": 5119940.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.754668188095093,
|
|
"epoch": 2.2526124488868695,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047614600485081657,
|
|
"loss": 5.4222,
|
|
"mean_token_accuracy": 0.1754487484693527,
|
|
"num_tokens": 5129789.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.7520379543304445,
|
|
"epoch": 2.2571558382553385,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047598738430632963,
|
|
"loss": 5.4475,
|
|
"mean_token_accuracy": 0.167902871966362,
|
|
"num_tokens": 5139570.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.700397777557373,
|
|
"epoch": 2.2616992276238075,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047582826784545146,
|
|
"loss": 5.4177,
|
|
"mean_token_accuracy": 0.1686418280005455,
|
|
"num_tokens": 5150526.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.7402173519134525,
|
|
"epoch": 2.266242616992276,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047566865586078615,
|
|
"loss": 5.3691,
|
|
"mean_token_accuracy": 0.1707841008901596,
|
|
"num_tokens": 5160895.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.727770662307739,
|
|
"epoch": 2.270786006360745,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047550854874616047,
|
|
"loss": 5.3712,
|
|
"mean_token_accuracy": 0.17329805642366408,
|
|
"num_tokens": 5171054.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.270786006360745,
|
|
"eval_entropy": 5.57227717111277,
|
|
"eval_loss": 5.902021408081055,
|
|
"eval_mean_token_accuracy": 0.1561654274720092,
|
|
"eval_num_tokens": 5171054.0,
|
|
"eval_runtime": 2.0014,
|
|
"eval_samples_per_second": 1716.327,
|
|
"eval_steps_per_second": 214.853,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.794205665588379,
|
|
"epoch": 2.275329395729214,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047534794689662257,
|
|
"loss": 5.5307,
|
|
"mean_token_accuracy": 0.15978193283081055,
|
|
"num_tokens": 5180870.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.847792863845825,
|
|
"epoch": 2.279872785097683,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004751868507084418,
|
|
"loss": 5.4595,
|
|
"mean_token_accuracy": 0.16842130720615386,
|
|
"num_tokens": 5190388.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.808324146270752,
|
|
"epoch": 2.284416174466152,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004750252605791068,
|
|
"loss": 5.4426,
|
|
"mean_token_accuracy": 0.16519180089235305,
|
|
"num_tokens": 5200582.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.752334976196289,
|
|
"epoch": 2.2889595638346205,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004748631769073254,
|
|
"loss": 5.4158,
|
|
"mean_token_accuracy": 0.17026202231645585,
|
|
"num_tokens": 5211079.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.790038681030273,
|
|
"epoch": 2.2935029532030895,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004747006000930227,
|
|
"loss": 5.5502,
|
|
"mean_token_accuracy": 0.16372255980968475,
|
|
"num_tokens": 5221481.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.804788255691529,
|
|
"epoch": 2.2980463425715585,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004745375305373409,
|
|
"loss": 5.4115,
|
|
"mean_token_accuracy": 0.17465941607952118,
|
|
"num_tokens": 5231387.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.644592189788819,
|
|
"epoch": 2.302589731940027,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047437396864263804,
|
|
"loss": 5.3736,
|
|
"mean_token_accuracy": 0.18058407604694365,
|
|
"num_tokens": 5240758.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.72665057182312,
|
|
"epoch": 2.307133121308496,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004742099148124866,
|
|
"loss": 5.3956,
|
|
"mean_token_accuracy": 0.17343889772892,
|
|
"num_tokens": 5250839.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.8157422065734865,
|
|
"epoch": 2.311676510676965,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004740453694516734,
|
|
"loss": 5.503,
|
|
"mean_token_accuracy": 0.1690310150384903,
|
|
"num_tokens": 5261554.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.7645652294158936,
|
|
"epoch": 2.316219900045434,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047388033296619746,
|
|
"loss": 5.4664,
|
|
"mean_token_accuracy": 0.17111266702413558,
|
|
"num_tokens": 5272422.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.743662118911743,
|
|
"epoch": 2.320763289413903,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004737148057632701,
|
|
"loss": 5.4025,
|
|
"mean_token_accuracy": 0.16985436230897905,
|
|
"num_tokens": 5282522.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.785736799240112,
|
|
"epoch": 2.3253066787823715,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004735487882513131,
|
|
"loss": 5.4442,
|
|
"mean_token_accuracy": 0.17556467056274414,
|
|
"num_tokens": 5293100.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.810127592086792,
|
|
"epoch": 2.3298500681508405,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004733822808399583,
|
|
"loss": 5.4832,
|
|
"mean_token_accuracy": 0.16948726028203964,
|
|
"num_tokens": 5303103.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.780827951431275,
|
|
"epoch": 2.3343934575193095,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047321528394004615,
|
|
"loss": 5.4578,
|
|
"mean_token_accuracy": 0.17639515399932862,
|
|
"num_tokens": 5313765.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.832022953033447,
|
|
"epoch": 2.3389368468877785,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047304779796362485,
|
|
"loss": 5.5494,
|
|
"mean_token_accuracy": 0.1650695815682411,
|
|
"num_tokens": 5324816.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.7504880905151365,
|
|
"epoch": 2.343480236256247,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047287982332394943,
|
|
"loss": 5.4324,
|
|
"mean_token_accuracy": 0.17428883910179138,
|
|
"num_tokens": 5334882.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.728056526184082,
|
|
"epoch": 2.348023625624716,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047271136043548057,
|
|
"loss": 5.422,
|
|
"mean_token_accuracy": 0.17303657233715058,
|
|
"num_tokens": 5344623.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.769865798950195,
|
|
"epoch": 2.352567014993185,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004725424097138838,
|
|
"loss": 5.4449,
|
|
"mean_token_accuracy": 0.16967813223600386,
|
|
"num_tokens": 5355786.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.6975971221923825,
|
|
"epoch": 2.357110404361654,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047237297157602816,
|
|
"loss": 5.436,
|
|
"mean_token_accuracy": 0.16897562593221666,
|
|
"num_tokens": 5366139.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.863748359680176,
|
|
"epoch": 2.3616537937301225,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004722030464399854,
|
|
"loss": 5.538,
|
|
"mean_token_accuracy": 0.16488790959119798,
|
|
"num_tokens": 5376890.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.7098846435546875,
|
|
"epoch": 2.3661971830985915,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000472032634725029,
|
|
"loss": 5.4449,
|
|
"mean_token_accuracy": 0.17615094780921936,
|
|
"num_tokens": 5386602.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.761862850189209,
|
|
"epoch": 2.3707405724670605,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047186173685163273,
|
|
"loss": 5.5115,
|
|
"mean_token_accuracy": 0.1679927960038185,
|
|
"num_tokens": 5397749.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.839893865585327,
|
|
"epoch": 2.3752839618355295,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047169035324147023,
|
|
"loss": 5.4687,
|
|
"mean_token_accuracy": 0.16668442338705064,
|
|
"num_tokens": 5407666.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.746318387985229,
|
|
"epoch": 2.379827351203998,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047151848431741363,
|
|
"loss": 5.5508,
|
|
"mean_token_accuracy": 0.16410186588764192,
|
|
"num_tokens": 5418668.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.706380319595337,
|
|
"epoch": 2.384370740572467,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004713461305035322,
|
|
"loss": 5.4168,
|
|
"mean_token_accuracy": 0.17993294596672058,
|
|
"num_tokens": 5429098.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.7328697681427006,
|
|
"epoch": 2.388914129940936,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000471173292225092,
|
|
"loss": 5.3732,
|
|
"mean_token_accuracy": 0.17878435105085372,
|
|
"num_tokens": 5438997.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.739597511291504,
|
|
"epoch": 2.393457519309405,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047099996990855424,
|
|
"loss": 5.4378,
|
|
"mean_token_accuracy": 0.17116021662950515,
|
|
"num_tokens": 5449315.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.728729867935181,
|
|
"epoch": 2.3980009086778735,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047082616398157456,
|
|
"loss": 5.4659,
|
|
"mean_token_accuracy": 0.1726785734295845,
|
|
"num_tokens": 5460523.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.834688806533814,
|
|
"epoch": 2.4025442980463425,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004706518748730017,
|
|
"loss": 5.5019,
|
|
"mean_token_accuracy": 0.16893154978752137,
|
|
"num_tokens": 5470012.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.713891077041626,
|
|
"epoch": 2.4070876874148115,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004704771030128768,
|
|
"loss": 5.4321,
|
|
"mean_token_accuracy": 0.1767338991165161,
|
|
"num_tokens": 5480364.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.838604497909546,
|
|
"epoch": 2.4116310767832805,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047030184883243203,
|
|
"loss": 5.5777,
|
|
"mean_token_accuracy": 0.1630810022354126,
|
|
"num_tokens": 5492138.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.729302930831909,
|
|
"epoch": 2.416174466151749,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047012611276408964,
|
|
"loss": 5.3506,
|
|
"mean_token_accuracy": 0.18401314169168473,
|
|
"num_tokens": 5501935.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.731316995620728,
|
|
"epoch": 2.420717855520218,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004699498952414609,
|
|
"loss": 5.5447,
|
|
"mean_token_accuracy": 0.16525724679231643,
|
|
"num_tokens": 5513993.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.778429746627808,
|
|
"epoch": 2.425261244888687,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046977319669934506,
|
|
"loss": 5.4661,
|
|
"mean_token_accuracy": 0.17474229335784913,
|
|
"num_tokens": 5524822.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.8102514266967775,
|
|
"epoch": 2.429804634257156,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046959601757372814,
|
|
"loss": 5.4488,
|
|
"mean_token_accuracy": 0.1700899377465248,
|
|
"num_tokens": 5535583.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.695425271987915,
|
|
"epoch": 2.4343480236256245,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046941835830178216,
|
|
"loss": 5.4298,
|
|
"mean_token_accuracy": 0.17716339081525803,
|
|
"num_tokens": 5546199.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.746195125579834,
|
|
"epoch": 2.4388914129940935,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046924021932186357,
|
|
"loss": 5.4054,
|
|
"mean_token_accuracy": 0.17828491628170012,
|
|
"num_tokens": 5556409.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.733344030380249,
|
|
"epoch": 2.4434348023625625,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046906160107351263,
|
|
"loss": 5.4656,
|
|
"mean_token_accuracy": 0.1751408025622368,
|
|
"num_tokens": 5565854.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.810946989059448,
|
|
"epoch": 2.4479781917310315,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00046888250399745217,
|
|
"loss": 5.4987,
|
|
"mean_token_accuracy": 0.163340325653553,
|
|
"num_tokens": 5576194.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.690517473220825,
|
|
"epoch": 2.4525215810995,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046870292853558646,
|
|
"loss": 5.4154,
|
|
"mean_token_accuracy": 0.17349067330360413,
|
|
"num_tokens": 5586679.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.691789197921753,
|
|
"epoch": 2.457064970467969,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046852287513099997,
|
|
"loss": 5.4446,
|
|
"mean_token_accuracy": 0.16185215711593628,
|
|
"num_tokens": 5597268.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.761885309219361,
|
|
"epoch": 2.461608359836438,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004683423442279566,
|
|
"loss": 5.4294,
|
|
"mean_token_accuracy": 0.17061847746372222,
|
|
"num_tokens": 5608046.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.756528043746949,
|
|
"epoch": 2.466151749204907,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004681613362718985,
|
|
"loss": 5.3423,
|
|
"mean_token_accuracy": 0.17276189625263214,
|
|
"num_tokens": 5617541.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.708715772628784,
|
|
"epoch": 2.470695138573376,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004679798517094447,
|
|
"loss": 5.4034,
|
|
"mean_token_accuracy": 0.17097639590501784,
|
|
"num_tokens": 5628428.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.7950883388519285,
|
|
"epoch": 2.4752385279418445,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046779789098839043,
|
|
"loss": 5.586,
|
|
"mean_token_accuracy": 0.1578267239034176,
|
|
"num_tokens": 5639557.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.850655841827392,
|
|
"epoch": 2.4797819173103135,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046761545455770555,
|
|
"loss": 5.5607,
|
|
"mean_token_accuracy": 0.17038990557193756,
|
|
"num_tokens": 5650586.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.7557775497436525,
|
|
"epoch": 2.4843253066787825,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004674325428675339,
|
|
"loss": 5.4627,
|
|
"mean_token_accuracy": 0.17183326482772826,
|
|
"num_tokens": 5661525.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.769146394729614,
|
|
"epoch": 2.488868696047251,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004672491563691919,
|
|
"loss": 5.4876,
|
|
"mean_token_accuracy": 0.1692006140947342,
|
|
"num_tokens": 5671272.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.751975154876709,
|
|
"epoch": 2.49341208541572,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046706529551516753,
|
|
"loss": 5.4049,
|
|
"mean_token_accuracy": 0.17520857751369476,
|
|
"num_tokens": 5680384.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.725416755676269,
|
|
"epoch": 2.497955474784189,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046688096075911905,
|
|
"loss": 5.4307,
|
|
"mean_token_accuracy": 0.1731862410902977,
|
|
"num_tokens": 5690983.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.693243312835693,
|
|
"epoch": 2.502498864152658,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004666961525558743,
|
|
"loss": 5.4599,
|
|
"mean_token_accuracy": 0.1667140856385231,
|
|
"num_tokens": 5702144.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.723059415817261,
|
|
"epoch": 2.507042253521127,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000466510871361429,
|
|
"loss": 5.5098,
|
|
"mean_token_accuracy": 0.16999765187501908,
|
|
"num_tokens": 5713283.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.744458770751953,
|
|
"epoch": 2.5115856428895955,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00046632511763294615,
|
|
"loss": 5.4728,
|
|
"mean_token_accuracy": 0.1633162885904312,
|
|
"num_tokens": 5724533.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.742016935348511,
|
|
"epoch": 2.5161290322580645,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004661388918287546,
|
|
"loss": 5.4302,
|
|
"mean_token_accuracy": 0.17126747071743012,
|
|
"num_tokens": 5735186.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.75839319229126,
|
|
"epoch": 2.5206724216265335,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004659521944083481,
|
|
"loss": 5.5014,
|
|
"mean_token_accuracy": 0.16637639999389647,
|
|
"num_tokens": 5746537.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.7699531555175785,
|
|
"epoch": 2.525215810995002,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004657650258323838,
|
|
"loss": 5.4586,
|
|
"mean_token_accuracy": 0.17205800563097,
|
|
"num_tokens": 5756666.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.782371664047242,
|
|
"epoch": 2.529759200363471,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046557738656268166,
|
|
"loss": 5.4878,
|
|
"mean_token_accuracy": 0.1685907945036888,
|
|
"num_tokens": 5766870.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.7292876720428465,
|
|
"epoch": 2.53430258973194,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000465389277062223,
|
|
"loss": 5.4662,
|
|
"mean_token_accuracy": 0.17555798590183258,
|
|
"num_tokens": 5776775.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.79610857963562,
|
|
"epoch": 2.538845979100409,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046520069779514917,
|
|
"loss": 5.4762,
|
|
"mean_token_accuracy": 0.1673141434788704,
|
|
"num_tokens": 5786993.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.690096712112426,
|
|
"epoch": 2.543389368468878,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004650116492267609,
|
|
"loss": 5.4619,
|
|
"mean_token_accuracy": 0.1756898507475853,
|
|
"num_tokens": 5796230.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.832179546356201,
|
|
"epoch": 2.5479327578373465,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004648221318235165,
|
|
"loss": 5.4503,
|
|
"mean_token_accuracy": 0.17338943034410476,
|
|
"num_tokens": 5807086.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.675332307815552,
|
|
"epoch": 2.5524761472058155,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004646321460530316,
|
|
"loss": 5.488,
|
|
"mean_token_accuracy": 0.16616813242435455,
|
|
"num_tokens": 5817897.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.789116477966308,
|
|
"epoch": 2.5570195365742845,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000464441692384077,
|
|
"loss": 5.3745,
|
|
"mean_token_accuracy": 0.18060711324214934,
|
|
"num_tokens": 5827794.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.6456574440002445,
|
|
"epoch": 2.561562925942753,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004642507712865784,
|
|
"loss": 5.4681,
|
|
"mean_token_accuracy": 0.17117729932069778,
|
|
"num_tokens": 5837603.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.695605659484864,
|
|
"epoch": 2.566106315311222,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046405938323161453,
|
|
"loss": 5.3799,
|
|
"mean_token_accuracy": 0.17985784262418747,
|
|
"num_tokens": 5847013.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.746594429016113,
|
|
"epoch": 2.570649704679691,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046386752869141624,
|
|
"loss": 5.3693,
|
|
"mean_token_accuracy": 0.17321225702762605,
|
|
"num_tokens": 5856783.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.699766159057617,
|
|
"epoch": 2.57519309404816,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004636752081393658,
|
|
"loss": 5.3794,
|
|
"mean_token_accuracy": 0.18014585226774216,
|
|
"num_tokens": 5866241.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.646414709091187,
|
|
"epoch": 2.579736483416629,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046348242204999487,
|
|
"loss": 5.4349,
|
|
"mean_token_accuracy": 0.17664633095264434,
|
|
"num_tokens": 5876135.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.74436583518982,
|
|
"epoch": 2.5842798727850975,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046328917089898396,
|
|
"loss": 5.4782,
|
|
"mean_token_accuracy": 0.1753312572836876,
|
|
"num_tokens": 5886841.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.7142030715942385,
|
|
"epoch": 2.5888232621535665,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004630954551631612,
|
|
"loss": 5.4249,
|
|
"mean_token_accuracy": 0.17433690130710602,
|
|
"num_tokens": 5896884.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.707109785079956,
|
|
"epoch": 2.5933666515220355,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00046290127532050087,
|
|
"loss": 5.4573,
|
|
"mean_token_accuracy": 0.17656322568655014,
|
|
"num_tokens": 5906194.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.715504121780396,
|
|
"epoch": 2.5979100408905045,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004627066318501223,
|
|
"loss": 5.3901,
|
|
"mean_token_accuracy": 0.18081092238426208,
|
|
"num_tokens": 5917217.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.662218999862671,
|
|
"epoch": 2.602453430258973,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046251152523228906,
|
|
"loss": 5.3173,
|
|
"mean_token_accuracy": 0.17789212614297867,
|
|
"num_tokens": 5926307.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.677583980560303,
|
|
"epoch": 2.606996819627442,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004623159559484074,
|
|
"loss": 5.4297,
|
|
"mean_token_accuracy": 0.17117145359516145,
|
|
"num_tokens": 5936718.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.658066177368164,
|
|
"epoch": 2.611540208995911,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004621199244810248,
|
|
"loss": 5.3929,
|
|
"mean_token_accuracy": 0.17647164762020112,
|
|
"num_tokens": 5947870.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.722538042068481,
|
|
"epoch": 2.61608359836438,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00046192343131382964,
|
|
"loss": 5.565,
|
|
"mean_token_accuracy": 0.17092299163341523,
|
|
"num_tokens": 5959090.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.811821174621582,
|
|
"epoch": 2.620626987732849,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046172647693164926,
|
|
"loss": 5.5328,
|
|
"mean_token_accuracy": 0.1675619050860405,
|
|
"num_tokens": 5970435.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.714699220657349,
|
|
"epoch": 2.6251703771013175,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004615290618204489,
|
|
"loss": 5.3555,
|
|
"mean_token_accuracy": 0.1846381425857544,
|
|
"num_tokens": 5979732.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.706509256362915,
|
|
"epoch": 2.6297137664697865,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004613311864673309,
|
|
"loss": 5.5057,
|
|
"mean_token_accuracy": 0.1713327094912529,
|
|
"num_tokens": 5989965.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.823319339752198,
|
|
"epoch": 2.6342571558382555,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046113285136053286,
|
|
"loss": 5.4775,
|
|
"mean_token_accuracy": 0.17892941683530808,
|
|
"num_tokens": 6000752.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.697203254699707,
|
|
"epoch": 2.638800545206724,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000460934056989427,
|
|
"loss": 5.4473,
|
|
"mean_token_accuracy": 0.17142748534679414,
|
|
"num_tokens": 6011318.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.688230037689209,
|
|
"epoch": 2.643343934575193,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004607348038445187,
|
|
"loss": 5.4467,
|
|
"mean_token_accuracy": 0.17383164465427398,
|
|
"num_tokens": 6022024.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.649292993545532,
|
|
"epoch": 2.647887323943662,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046053509241744526,
|
|
"loss": 5.4088,
|
|
"mean_token_accuracy": 0.17349524945020675,
|
|
"num_tokens": 6032118.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.755839061737061,
|
|
"epoch": 2.652430713312131,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046033492320097483,
|
|
"loss": 5.4266,
|
|
"mean_token_accuracy": 0.1789483055472374,
|
|
"num_tokens": 6042607.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.6492063999176025,
|
|
"epoch": 2.6569741026806,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004601342966890049,
|
|
"loss": 5.4267,
|
|
"mean_token_accuracy": 0.1725103497505188,
|
|
"num_tokens": 6053625.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.719505977630615,
|
|
"epoch": 2.6615174920490685,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045993321337656163,
|
|
"loss": 5.3997,
|
|
"mean_token_accuracy": 0.17744006365537643,
|
|
"num_tokens": 6063823.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.716916704177857,
|
|
"epoch": 2.6660608814175375,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000459731673759798,
|
|
"loss": 5.3951,
|
|
"mean_token_accuracy": 0.1731013387441635,
|
|
"num_tokens": 6073914.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.714711332321167,
|
|
"epoch": 2.6706042707860065,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045952967833599303,
|
|
"loss": 5.3769,
|
|
"mean_token_accuracy": 0.1751635730266571,
|
|
"num_tokens": 6085121.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.629806327819824,
|
|
"epoch": 2.675147660154475,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004593272276035502,
|
|
"loss": 5.3008,
|
|
"mean_token_accuracy": 0.18416397869586945,
|
|
"num_tokens": 6095324.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.652683973312378,
|
|
"epoch": 2.679691049522944,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004591243220619969,
|
|
"loss": 5.4307,
|
|
"mean_token_accuracy": 0.17971840053796767,
|
|
"num_tokens": 6104704.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.7351244449615475,
|
|
"epoch": 2.684234438891413,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000458920962211982,
|
|
"loss": 5.5148,
|
|
"mean_token_accuracy": 0.17182195484638213,
|
|
"num_tokens": 6116569.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.812714910507202,
|
|
"epoch": 2.688777828259882,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004587171485552759,
|
|
"loss": 5.5145,
|
|
"mean_token_accuracy": 0.1733253166079521,
|
|
"num_tokens": 6126166.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.702657794952392,
|
|
"epoch": 2.693321217628351,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045851288159476854,
|
|
"loss": 5.3752,
|
|
"mean_token_accuracy": 0.17850163578987122,
|
|
"num_tokens": 6136169.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.766521549224853,
|
|
"epoch": 2.6978646069968195,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004583081618344684,
|
|
"loss": 5.4473,
|
|
"mean_token_accuracy": 0.17969665676355362,
|
|
"num_tokens": 6146249.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.823341035842896,
|
|
"epoch": 2.7024079963652885,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000458102989779501,
|
|
"loss": 5.5272,
|
|
"mean_token_accuracy": 0.17510570585727692,
|
|
"num_tokens": 6156892.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.735653495788574,
|
|
"epoch": 2.7069513857337575,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004578973659361081,
|
|
"loss": 5.3965,
|
|
"mean_token_accuracy": 0.18579644560813904,
|
|
"num_tokens": 6166915.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.687149286270142,
|
|
"epoch": 2.711494775102226,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045769129081164596,
|
|
"loss": 5.3385,
|
|
"mean_token_accuracy": 0.1777852803468704,
|
|
"num_tokens": 6176837.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.582538032531739,
|
|
"epoch": 2.716038164470695,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004574847649145845,
|
|
"loss": 5.4347,
|
|
"mean_token_accuracy": 0.17326221764087676,
|
|
"num_tokens": 6188038.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.728745841979981,
|
|
"epoch": 2.720581553839164,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000457277788754506,
|
|
"loss": 5.4644,
|
|
"mean_token_accuracy": 0.17362224608659743,
|
|
"num_tokens": 6199835.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.773366022109985,
|
|
"epoch": 2.725124943207633,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045707036284210344,
|
|
"loss": 5.3775,
|
|
"mean_token_accuracy": 0.1746920347213745,
|
|
"num_tokens": 6209122.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.725124943207633,
|
|
"eval_entropy": 5.490004498459572,
|
|
"eval_loss": 5.812583923339844,
|
|
"eval_mean_token_accuracy": 0.16149513153142708,
|
|
"eval_num_tokens": 6209122.0,
|
|
"eval_runtime": 2.0111,
|
|
"eval_samples_per_second": 1708.014,
|
|
"eval_steps_per_second": 213.813,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.657353448867798,
|
|
"epoch": 2.729668332576102,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004568624876891796,
|
|
"loss": 5.4355,
|
|
"mean_token_accuracy": 0.17501228600740432,
|
|
"num_tokens": 6219317.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.745296287536621,
|
|
"epoch": 2.7342117219445705,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045665416380864587,
|
|
"loss": 5.4687,
|
|
"mean_token_accuracy": 0.1732914850115776,
|
|
"num_tokens": 6229561.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.717855453491211,
|
|
"epoch": 2.7387551113130395,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004564453917145206,
|
|
"loss": 5.4435,
|
|
"mean_token_accuracy": 0.17248283475637435,
|
|
"num_tokens": 6240150.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.660070943832397,
|
|
"epoch": 2.7432985006815085,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004562361719219283,
|
|
"loss": 5.3201,
|
|
"mean_token_accuracy": 0.17407702952623366,
|
|
"num_tokens": 6250641.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.658320045471191,
|
|
"epoch": 2.747841890049977,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045602650494709786,
|
|
"loss": 5.4327,
|
|
"mean_token_accuracy": 0.17014171481132506,
|
|
"num_tokens": 6261239.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.726068162918091,
|
|
"epoch": 2.752385279418446,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00045581639130736173,
|
|
"loss": 5.4294,
|
|
"mean_token_accuracy": 0.17162348777055741,
|
|
"num_tokens": 6270919.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.71975417137146,
|
|
"epoch": 2.756928668786915,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004556058315211546,
|
|
"loss": 5.5011,
|
|
"mean_token_accuracy": 0.16653174459934234,
|
|
"num_tokens": 6283007.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.5943902969360355,
|
|
"epoch": 2.761472058155384,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045539482610801156,
|
|
"loss": 5.3545,
|
|
"mean_token_accuracy": 0.17958497405052185,
|
|
"num_tokens": 6293596.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.725016927719116,
|
|
"epoch": 2.766015447523853,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004551833755885677,
|
|
"loss": 5.4087,
|
|
"mean_token_accuracy": 0.17834171205759047,
|
|
"num_tokens": 6304568.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.672103452682495,
|
|
"epoch": 2.7705588368923215,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004549714804845561,
|
|
"loss": 5.3788,
|
|
"mean_token_accuracy": 0.17516253739595414,
|
|
"num_tokens": 6315275.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.7183074951171875,
|
|
"epoch": 2.7751022262607905,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004547591413188068,
|
|
"loss": 5.4891,
|
|
"mean_token_accuracy": 0.17295058369636535,
|
|
"num_tokens": 6324986.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.72510347366333,
|
|
"epoch": 2.7796456156292595,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004545463586152456,
|
|
"loss": 5.4776,
|
|
"mean_token_accuracy": 0.16752664595842362,
|
|
"num_tokens": 6336318.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.638709497451782,
|
|
"epoch": 2.784189004997728,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004543331328988927,
|
|
"loss": 5.3658,
|
|
"mean_token_accuracy": 0.1809811234474182,
|
|
"num_tokens": 6345986.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.683207178115845,
|
|
"epoch": 2.788732394366197,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004541194646958614,
|
|
"loss": 5.4427,
|
|
"mean_token_accuracy": 0.17370019406080245,
|
|
"num_tokens": 6357025.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.748136377334594,
|
|
"epoch": 2.793275783734666,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045390535453335666,
|
|
"loss": 5.4465,
|
|
"mean_token_accuracy": 0.17309835702180862,
|
|
"num_tokens": 6367885.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.555459022521973,
|
|
"epoch": 2.797819173103135,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045369080293967414,
|
|
"loss": 5.3708,
|
|
"mean_token_accuracy": 0.17574317902326583,
|
|
"num_tokens": 6377729.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.659260272979736,
|
|
"epoch": 2.802362562471604,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045347581044419846,
|
|
"loss": 5.3641,
|
|
"mean_token_accuracy": 0.18111564815044404,
|
|
"num_tokens": 6387368.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.751581239700317,
|
|
"epoch": 2.806905951840073,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045326037757740216,
|
|
"loss": 5.4795,
|
|
"mean_token_accuracy": 0.17150890678167344,
|
|
"num_tokens": 6397771.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.694161748886108,
|
|
"epoch": 2.8114493412085415,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004530445048708447,
|
|
"loss": 5.3471,
|
|
"mean_token_accuracy": 0.18445285856723787,
|
|
"num_tokens": 6408238.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.7077723979949955,
|
|
"epoch": 2.8159927305770105,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045282819285717024,
|
|
"loss": 5.4931,
|
|
"mean_token_accuracy": 0.1682308092713356,
|
|
"num_tokens": 6419196.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.67715220451355,
|
|
"epoch": 2.8205361199454795,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00045261144207010736,
|
|
"loss": 5.4401,
|
|
"mean_token_accuracy": 0.17278133779764177,
|
|
"num_tokens": 6430342.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.702464771270752,
|
|
"epoch": 2.825079509313948,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045239425304446684,
|
|
"loss": 5.3351,
|
|
"mean_token_accuracy": 0.18193444311618806,
|
|
"num_tokens": 6440218.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.645659351348877,
|
|
"epoch": 2.829622898682417,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004521766263161413,
|
|
"loss": 5.3341,
|
|
"mean_token_accuracy": 0.18401312828063965,
|
|
"num_tokens": 6450128.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.614056396484375,
|
|
"epoch": 2.834166288050886,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000451958562422103,
|
|
"loss": 5.3534,
|
|
"mean_token_accuracy": 0.18784762024879456,
|
|
"num_tokens": 6459916.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.729891920089722,
|
|
"epoch": 2.838709677419355,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045174006190040267,
|
|
"loss": 5.3877,
|
|
"mean_token_accuracy": 0.17789144217967987,
|
|
"num_tokens": 6470250.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.720960330963135,
|
|
"epoch": 2.843253066787824,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000451521125290169,
|
|
"loss": 5.4833,
|
|
"mean_token_accuracy": 0.1697496220469475,
|
|
"num_tokens": 6481623.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.6947548389434814,
|
|
"epoch": 2.8477964561562925,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000451301753131606,
|
|
"loss": 5.3849,
|
|
"mean_token_accuracy": 0.17648440152406691,
|
|
"num_tokens": 6492143.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.711607503890991,
|
|
"epoch": 2.8523398455247615,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045108194596599286,
|
|
"loss": 5.4067,
|
|
"mean_token_accuracy": 0.18158916085958482,
|
|
"num_tokens": 6501643.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.620377874374389,
|
|
"epoch": 2.8568832348932305,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004508617043356818,
|
|
"loss": 5.337,
|
|
"mean_token_accuracy": 0.18244485408067704,
|
|
"num_tokens": 6511442.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.621693849563599,
|
|
"epoch": 2.861426624261699,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004506410287840972,
|
|
"loss": 5.2693,
|
|
"mean_token_accuracy": 0.18329064995050431,
|
|
"num_tokens": 6521303.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.5911907196044925,
|
|
"epoch": 2.865970013630168,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000450419919855734,
|
|
"loss": 5.4053,
|
|
"mean_token_accuracy": 0.17518810778856278,
|
|
"num_tokens": 6531143.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.718419027328491,
|
|
"epoch": 2.870513402998637,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004501983780961566,
|
|
"loss": 5.4538,
|
|
"mean_token_accuracy": 0.17734911143779755,
|
|
"num_tokens": 6540899.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.807937812805176,
|
|
"epoch": 2.875056792367106,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004499764040519972,
|
|
"loss": 5.5017,
|
|
"mean_token_accuracy": 0.17058033794164656,
|
|
"num_tokens": 6552131.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.657841634750366,
|
|
"epoch": 2.879600181735575,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044975399827095467,
|
|
"loss": 5.4311,
|
|
"mean_token_accuracy": 0.17171088755130767,
|
|
"num_tokens": 6563275.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.67250280380249,
|
|
"epoch": 2.8841435711040435,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044953116130179313,
|
|
"loss": 5.2791,
|
|
"mean_token_accuracy": 0.18332293778657913,
|
|
"num_tokens": 6573109.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.600630617141723,
|
|
"epoch": 2.8886869604725125,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004493078936943408,
|
|
"loss": 5.4135,
|
|
"mean_token_accuracy": 0.16953012347221375,
|
|
"num_tokens": 6584978.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.631850290298462,
|
|
"epoch": 2.8932303498409815,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004490841959994882,
|
|
"loss": 5.4107,
|
|
"mean_token_accuracy": 0.17511872500181197,
|
|
"num_tokens": 6595608.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.61538872718811,
|
|
"epoch": 2.89777373920945,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044886006876918716,
|
|
"loss": 5.2887,
|
|
"mean_token_accuracy": 0.17951602339744568,
|
|
"num_tokens": 6605983.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.685514116287232,
|
|
"epoch": 2.902317128577919,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004486355125564493,
|
|
"loss": 5.4145,
|
|
"mean_token_accuracy": 0.17293050438165664,
|
|
"num_tokens": 6616682.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.745040988922119,
|
|
"epoch": 2.906860517946388,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004484105279153449,
|
|
"loss": 5.4559,
|
|
"mean_token_accuracy": 0.17348510175943374,
|
|
"num_tokens": 6628715.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.701951503753662,
|
|
"epoch": 2.911403907314857,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000448185115401001,
|
|
"loss": 5.446,
|
|
"mean_token_accuracy": 0.17357207983732223,
|
|
"num_tokens": 6639219.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.640505695343018,
|
|
"epoch": 2.915947296683326,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004479592755696008,
|
|
"loss": 5.3164,
|
|
"mean_token_accuracy": 0.18428571224212648,
|
|
"num_tokens": 6649516.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.685490131378174,
|
|
"epoch": 2.9204906860517945,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044773300897838137,
|
|
"loss": 5.4374,
|
|
"mean_token_accuracy": 0.17735814154148102,
|
|
"num_tokens": 6659470.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.639790105819702,
|
|
"epoch": 2.9250340754202635,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004475063161856332,
|
|
"loss": 5.4132,
|
|
"mean_token_accuracy": 0.1750086635351181,
|
|
"num_tokens": 6669637.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.623899269104004,
|
|
"epoch": 2.9295774647887325,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004472791977506983,
|
|
"loss": 5.3596,
|
|
"mean_token_accuracy": 0.17347195148468017,
|
|
"num_tokens": 6680450.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.652746677398682,
|
|
"epoch": 2.934120854157201,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044705165423396855,
|
|
"loss": 5.4471,
|
|
"mean_token_accuracy": 0.17181625962257385,
|
|
"num_tokens": 6691526.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.708565092086792,
|
|
"epoch": 2.93866424352567,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00044682368619688523,
|
|
"loss": 5.3877,
|
|
"mean_token_accuracy": 0.18078314810991286,
|
|
"num_tokens": 6702543.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.660660934448242,
|
|
"epoch": 2.943207632894139,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004465952942019367,
|
|
"loss": 5.4692,
|
|
"mean_token_accuracy": 0.17159684151411056,
|
|
"num_tokens": 6712374.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.67660665512085,
|
|
"epoch": 2.947751022262608,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044636647881265745,
|
|
"loss": 5.36,
|
|
"mean_token_accuracy": 0.17741062194108964,
|
|
"num_tokens": 6722091.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.610265445709229,
|
|
"epoch": 2.952294411631077,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004461372405936269,
|
|
"loss": 5.3495,
|
|
"mean_token_accuracy": 0.18504534810781478,
|
|
"num_tokens": 6732024.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.683926057815552,
|
|
"epoch": 2.9568378009995455,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004459075801104673,
|
|
"loss": 5.4683,
|
|
"mean_token_accuracy": 0.17421613335609437,
|
|
"num_tokens": 6741480.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.725023078918457,
|
|
"epoch": 2.9613811903680145,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00044567749792984325,
|
|
"loss": 5.3408,
|
|
"mean_token_accuracy": 0.17926502376794815,
|
|
"num_tokens": 6752504.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.6624329566955565,
|
|
"epoch": 2.9659245797364835,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044544699461945966,
|
|
"loss": 5.3922,
|
|
"mean_token_accuracy": 0.17102901488542557,
|
|
"num_tokens": 6762945.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.634073781967163,
|
|
"epoch": 2.970467969104952,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044521607074806037,
|
|
"loss": 5.3687,
|
|
"mean_token_accuracy": 0.17886101454496384,
|
|
"num_tokens": 6772862.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.723052597045898,
|
|
"epoch": 2.975011358473421,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004449847268854272,
|
|
"loss": 5.3955,
|
|
"mean_token_accuracy": 0.1785761758685112,
|
|
"num_tokens": 6782912.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.695500755310059,
|
|
"epoch": 2.97955474784189,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004447529636023782,
|
|
"loss": 5.4192,
|
|
"mean_token_accuracy": 0.1772427439689636,
|
|
"num_tokens": 6793249.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.659029817581176,
|
|
"epoch": 2.984098137210359,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044452078147076614,
|
|
"loss": 5.4184,
|
|
"mean_token_accuracy": 0.17817537039518355,
|
|
"num_tokens": 6803721.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.762314653396606,
|
|
"epoch": 2.988641526578828,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004442881810634774,
|
|
"loss": 5.4045,
|
|
"mean_token_accuracy": 0.17852256894111634,
|
|
"num_tokens": 6813637.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.676381158828735,
|
|
"epoch": 2.9931849159472965,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004440551629544303,
|
|
"loss": 5.3922,
|
|
"mean_token_accuracy": 0.1706793874502182,
|
|
"num_tokens": 6824171.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.5360705852508545,
|
|
"epoch": 2.9977283053157655,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044382172771857383,
|
|
"loss": 5.2542,
|
|
"mean_token_accuracy": 0.1853314682841301,
|
|
"num_tokens": 6833668.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.615872224171956,
|
|
"epoch": 3.0018173557473875,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004435878759318864,
|
|
"loss": 5.1763,
|
|
"mean_token_accuracy": 0.18854615092277527,
|
|
"num_tokens": 6842297.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.712144136428833,
|
|
"epoch": 3.0063607451158565,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004433536081713738,
|
|
"loss": 5.174,
|
|
"mean_token_accuracy": 0.18712559342384338,
|
|
"num_tokens": 6853124.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.612299823760987,
|
|
"epoch": 3.0109041344843255,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004431189250150686,
|
|
"loss": 5.0484,
|
|
"mean_token_accuracy": 0.1988249495625496,
|
|
"num_tokens": 6861990.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.643168306350708,
|
|
"epoch": 3.015447523852794,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044288382704202816,
|
|
"loss": 5.0479,
|
|
"mean_token_accuracy": 0.19690826982259751,
|
|
"num_tokens": 6872024.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.613150024414063,
|
|
"epoch": 3.019990913221263,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044264831483233335,
|
|
"loss": 5.1094,
|
|
"mean_token_accuracy": 0.19287334233522416,
|
|
"num_tokens": 6881205.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.536692142486572,
|
|
"epoch": 3.024534302589732,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044241238896708716,
|
|
"loss": 5.0683,
|
|
"mean_token_accuracy": 0.19721028953790665,
|
|
"num_tokens": 6891741.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.530151414871216,
|
|
"epoch": 3.029077691958201,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004421760500284131,
|
|
"loss": 5.1507,
|
|
"mean_token_accuracy": 0.18868669271469116,
|
|
"num_tokens": 6902122.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.651425218582153,
|
|
"epoch": 3.0336210813266695,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004419392985994541,
|
|
"loss": 5.1042,
|
|
"mean_token_accuracy": 0.19284143000841142,
|
|
"num_tokens": 6912129.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.529677820205689,
|
|
"epoch": 3.0381644706951385,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044170213526437076,
|
|
"loss": 4.9918,
|
|
"mean_token_accuracy": 0.20241980701684953,
|
|
"num_tokens": 6921409.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.553798770904541,
|
|
"epoch": 3.0427078600636075,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044146456060833994,
|
|
"loss": 5.1186,
|
|
"mean_token_accuracy": 0.19057913571596147,
|
|
"num_tokens": 6932346.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.65966420173645,
|
|
"epoch": 3.0472512494320765,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044122657521755373,
|
|
"loss": 5.1213,
|
|
"mean_token_accuracy": 0.19494247138500215,
|
|
"num_tokens": 6942139.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.58785400390625,
|
|
"epoch": 3.051794638800545,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044098817967921716,
|
|
"loss": 5.1252,
|
|
"mean_token_accuracy": 0.191215680539608,
|
|
"num_tokens": 6952344.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.629664516448974,
|
|
"epoch": 3.056338028169014,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044074937458154763,
|
|
"loss": 5.1186,
|
|
"mean_token_accuracy": 0.1954929158091545,
|
|
"num_tokens": 6962430.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.626881647109985,
|
|
"epoch": 3.060881417537483,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00044051016051377303,
|
|
"loss": 5.1703,
|
|
"mean_token_accuracy": 0.18877767622470856,
|
|
"num_tokens": 6973213.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.57646074295044,
|
|
"epoch": 3.065424806905952,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004402705380661302,
|
|
"loss": 5.1081,
|
|
"mean_token_accuracy": 0.1910523384809494,
|
|
"num_tokens": 6983896.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.605443143844605,
|
|
"epoch": 3.0699681962744205,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044003050782986377,
|
|
"loss": 5.0766,
|
|
"mean_token_accuracy": 0.20163127779960632,
|
|
"num_tokens": 6993549.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.542435550689698,
|
|
"epoch": 3.0745115856428895,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004397900703972245,
|
|
"loss": 5.1059,
|
|
"mean_token_accuracy": 0.19292509704828262,
|
|
"num_tokens": 7003893.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.6184694290161135,
|
|
"epoch": 3.0790549750113585,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004395492263614677,
|
|
"loss": 5.2096,
|
|
"mean_token_accuracy": 0.18556181341409683,
|
|
"num_tokens": 7014826.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.696637678146362,
|
|
"epoch": 3.0835983643798275,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043930797631685234,
|
|
"loss": 5.1457,
|
|
"mean_token_accuracy": 0.19268843829631804,
|
|
"num_tokens": 7024758.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.550245046615601,
|
|
"epoch": 3.088141753748296,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043906632085863876,
|
|
"loss": 5.1018,
|
|
"mean_token_accuracy": 0.20011179596185685,
|
|
"num_tokens": 7034063.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.543308544158935,
|
|
"epoch": 3.092685143116765,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00043882426058308775,
|
|
"loss": 5.1119,
|
|
"mean_token_accuracy": 0.18908465504646302,
|
|
"num_tokens": 7045137.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.487951755523682,
|
|
"epoch": 3.097228532485234,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00043858179608745915,
|
|
"loss": 5.0506,
|
|
"mean_token_accuracy": 0.19407066106796264,
|
|
"num_tokens": 7056381.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.6910719871521,
|
|
"epoch": 3.101771921853703,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004383389279700098,
|
|
"loss": 5.17,
|
|
"mean_token_accuracy": 0.18348209410905839,
|
|
"num_tokens": 7066462.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.680981016159057,
|
|
"epoch": 3.1063153112221715,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043809565682999274,
|
|
"loss": 5.2062,
|
|
"mean_token_accuracy": 0.18687530755996704,
|
|
"num_tokens": 7078438.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.529812335968018,
|
|
"epoch": 3.1108587005906405,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043785198326765535,
|
|
"loss": 5.0606,
|
|
"mean_token_accuracy": 0.19707969427108765,
|
|
"num_tokens": 7088981.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.591468238830567,
|
|
"epoch": 3.1154020899591095,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043760790788423785,
|
|
"loss": 5.1141,
|
|
"mean_token_accuracy": 0.19358151257038117,
|
|
"num_tokens": 7099407.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.584726810455322,
|
|
"epoch": 3.1199454793275785,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043736343128197207,
|
|
"loss": 5.0303,
|
|
"mean_token_accuracy": 0.2007886588573456,
|
|
"num_tokens": 7109454.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.475041723251342,
|
|
"epoch": 3.124488868696047,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004371185540640796,
|
|
"loss": 5.1217,
|
|
"mean_token_accuracy": 0.18972404897212983,
|
|
"num_tokens": 7119639.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.604704856872559,
|
|
"epoch": 3.129032258064516,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043687327683477077,
|
|
"loss": 5.0772,
|
|
"mean_token_accuracy": 0.1982833191752434,
|
|
"num_tokens": 7129872.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.525883340835572,
|
|
"epoch": 3.133575647432985,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004366276001992427,
|
|
"loss": 5.0309,
|
|
"mean_token_accuracy": 0.20182539373636246,
|
|
"num_tokens": 7140456.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.481144523620605,
|
|
"epoch": 3.138119036801454,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000436381524763678,
|
|
"loss": 5.0672,
|
|
"mean_token_accuracy": 0.19743612557649612,
|
|
"num_tokens": 7149787.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.533722543716431,
|
|
"epoch": 3.142662426169923,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004361350511352435,
|
|
"loss": 5.0352,
|
|
"mean_token_accuracy": 0.19982903450727463,
|
|
"num_tokens": 7160482.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.589895915985108,
|
|
"epoch": 3.1472058155383915,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004358881799220882,
|
|
"loss": 5.0845,
|
|
"mean_token_accuracy": 0.19687338471412658,
|
|
"num_tokens": 7170818.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.5270490646362305,
|
|
"epoch": 3.1517492049068605,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00043564091173334236,
|
|
"loss": 5.1195,
|
|
"mean_token_accuracy": 0.1979329138994217,
|
|
"num_tokens": 7180770.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.492447996139527,
|
|
"epoch": 3.1562925942753295,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00043539324717911573,
|
|
"loss": 5.0302,
|
|
"mean_token_accuracy": 0.2038204178214073,
|
|
"num_tokens": 7190751.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.569909906387329,
|
|
"epoch": 3.1608359836437985,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00043514518687049587,
|
|
"loss": 5.044,
|
|
"mean_token_accuracy": 0.20199652314186095,
|
|
"num_tokens": 7200093.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.627804613113403,
|
|
"epoch": 3.165379373012267,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000434896731419547,
|
|
"loss": 5.1504,
|
|
"mean_token_accuracy": 0.19484904706478118,
|
|
"num_tokens": 7210526.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.554333448410034,
|
|
"epoch": 3.169922762380736,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004346478814393082,
|
|
"loss": 5.097,
|
|
"mean_token_accuracy": 0.19816578328609466,
|
|
"num_tokens": 7222127.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.484754657745361,
|
|
"epoch": 3.174466151749205,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004343986375437921,
|
|
"loss": 5.057,
|
|
"mean_token_accuracy": 0.2039112001657486,
|
|
"num_tokens": 7233418.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.534359264373779,
|
|
"epoch": 3.179009541117674,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043414900034798314,
|
|
"loss": 5.1329,
|
|
"mean_token_accuracy": 0.1940907284617424,
|
|
"num_tokens": 7243743.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.179009541117674,
|
|
"eval_entropy": 5.310462300721989,
|
|
"eval_loss": 5.785341739654541,
|
|
"eval_mean_token_accuracy": 0.1645617822748284,
|
|
"eval_num_tokens": 7243743.0,
|
|
"eval_runtime": 2.0051,
|
|
"eval_samples_per_second": 1713.152,
|
|
"eval_steps_per_second": 214.456,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.5337999820709225,
|
|
"epoch": 3.1835529304861425,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004338989704678364,
|
|
"loss": 5.1602,
|
|
"mean_token_accuracy": 0.18631302416324616,
|
|
"num_tokens": 7254906.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.594093751907349,
|
|
"epoch": 3.1880963198546115,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00043364854852027566,
|
|
"loss": 5.0995,
|
|
"mean_token_accuracy": 0.1928297445178032,
|
|
"num_tokens": 7264193.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.587867164611817,
|
|
"epoch": 3.1926397092230805,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004333977351231924,
|
|
"loss": 5.1409,
|
|
"mean_token_accuracy": 0.1957073375582695,
|
|
"num_tokens": 7273723.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.569634294509887,
|
|
"epoch": 3.1971830985915495,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043314653089544336,
|
|
"loss": 5.1783,
|
|
"mean_token_accuracy": 0.18770786076784135,
|
|
"num_tokens": 7284142.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.554693126678467,
|
|
"epoch": 3.201726487960018,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004328949364568504,
|
|
"loss": 5.0718,
|
|
"mean_token_accuracy": 0.1964555189013481,
|
|
"num_tokens": 7294228.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.546922588348389,
|
|
"epoch": 3.206269877328487,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004326429524281975,
|
|
"loss": 5.0996,
|
|
"mean_token_accuracy": 0.19577673971652984,
|
|
"num_tokens": 7304555.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.583894824981689,
|
|
"epoch": 3.210813266696956,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004323905794312303,
|
|
"loss": 5.1981,
|
|
"mean_token_accuracy": 0.18419464826583862,
|
|
"num_tokens": 7315955.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.529076957702637,
|
|
"epoch": 3.215356656065425,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00043213781808865414,
|
|
"loss": 5.1196,
|
|
"mean_token_accuracy": 0.18837847262620927,
|
|
"num_tokens": 7326757.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.5427507877349855,
|
|
"epoch": 3.2199000454338935,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004318846690241324,
|
|
"loss": 5.0913,
|
|
"mean_token_accuracy": 0.20070978701114656,
|
|
"num_tokens": 7335969.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.4962132453918455,
|
|
"epoch": 3.2244434348023625,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00043163113286228533,
|
|
"loss": 5.0987,
|
|
"mean_token_accuracy": 0.19617622047662736,
|
|
"num_tokens": 7345824.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.600872802734375,
|
|
"epoch": 3.2289868241708315,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004313772102286882,
|
|
"loss": 5.1789,
|
|
"mean_token_accuracy": 0.18539532124996186,
|
|
"num_tokens": 7356292.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.584480667114258,
|
|
"epoch": 3.2335302135393005,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004311229017498698,
|
|
"loss": 5.1675,
|
|
"mean_token_accuracy": 0.1897214397788048,
|
|
"num_tokens": 7366809.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.577239894866944,
|
|
"epoch": 3.238073602907769,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000430868208053311,
|
|
"loss": 5.0849,
|
|
"mean_token_accuracy": 0.2046733319759369,
|
|
"num_tokens": 7376076.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.543262958526611,
|
|
"epoch": 3.242616992276238,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00043061312976744326,
|
|
"loss": 5.106,
|
|
"mean_token_accuracy": 0.1914639040827751,
|
|
"num_tokens": 7386338.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.5331192970275875,
|
|
"epoch": 3.247160381644707,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004303576675216469,
|
|
"loss": 5.0926,
|
|
"mean_token_accuracy": 0.19141370952129363,
|
|
"num_tokens": 7395792.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.55092568397522,
|
|
"epoch": 3.251703771013176,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043010182194624957,
|
|
"loss": 5.1093,
|
|
"mean_token_accuracy": 0.19484492242336274,
|
|
"num_tokens": 7405765.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.503778028488159,
|
|
"epoch": 3.2562471603816445,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004298455936725248,
|
|
"loss": 5.1518,
|
|
"mean_token_accuracy": 0.19049873501062392,
|
|
"num_tokens": 7415213.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.624447298049927,
|
|
"epoch": 3.2607905497501135,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004295889833326903,
|
|
"loss": 5.1341,
|
|
"mean_token_accuracy": 0.18939176350831985,
|
|
"num_tokens": 7425722.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.640601253509521,
|
|
"epoch": 3.2653339391185825,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042933199155990675,
|
|
"loss": 5.1827,
|
|
"mean_token_accuracy": 0.1922486960887909,
|
|
"num_tokens": 7436240.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.542389917373657,
|
|
"epoch": 3.2698773284870515,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004290746189882757,
|
|
"loss": 5.1507,
|
|
"mean_token_accuracy": 0.18639931231737136,
|
|
"num_tokens": 7447878.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.577819728851319,
|
|
"epoch": 3.27442071785552,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00042881686625283844,
|
|
"loss": 5.1323,
|
|
"mean_token_accuracy": 0.19540536850690843,
|
|
"num_tokens": 7458635.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.5815582275390625,
|
|
"epoch": 3.278964107223989,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004285587339895742,
|
|
"loss": 5.1636,
|
|
"mean_token_accuracy": 0.18532266467809677,
|
|
"num_tokens": 7468740.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.665569591522217,
|
|
"epoch": 3.283507496592458,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042830022283539863,
|
|
"loss": 5.2204,
|
|
"mean_token_accuracy": 0.1837853506207466,
|
|
"num_tokens": 7480977.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.5470365524292,
|
|
"epoch": 3.288050885960927,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004280413334281623,
|
|
"loss": 5.2089,
|
|
"mean_token_accuracy": 0.18447837382555007,
|
|
"num_tokens": 7491758.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.592721939086914,
|
|
"epoch": 3.292594275329396,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00042778206640664915,
|
|
"loss": 5.1506,
|
|
"mean_token_accuracy": 0.18573983609676362,
|
|
"num_tokens": 7500486.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.489808988571167,
|
|
"epoch": 3.2971376646978645,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004275224224105748,
|
|
"loss": 5.1432,
|
|
"mean_token_accuracy": 0.19605191946029663,
|
|
"num_tokens": 7510371.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.566692066192627,
|
|
"epoch": 3.3016810540663335,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042726240208058484,
|
|
"loss": 5.1693,
|
|
"mean_token_accuracy": 0.196818944811821,
|
|
"num_tokens": 7520697.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.557655572891235,
|
|
"epoch": 3.3062244434348025,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004270020060582538,
|
|
"loss": 5.0877,
|
|
"mean_token_accuracy": 0.19443976283073425,
|
|
"num_tokens": 7530757.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.552946996688843,
|
|
"epoch": 3.310767832803271,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00042674123498608285,
|
|
"loss": 5.1219,
|
|
"mean_token_accuracy": 0.1895186722278595,
|
|
"num_tokens": 7541810.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.624138927459716,
|
|
"epoch": 3.31531122217174,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042648008950749877,
|
|
"loss": 5.3155,
|
|
"mean_token_accuracy": 0.18076227307319642,
|
|
"num_tokens": 7552754.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.604369878768921,
|
|
"epoch": 3.319854611540209,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000426218570266852,
|
|
"loss": 5.1006,
|
|
"mean_token_accuracy": 0.19819563776254653,
|
|
"num_tokens": 7562906.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.434622430801392,
|
|
"epoch": 3.324398000908678,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042595667790941536,
|
|
"loss": 5.1142,
|
|
"mean_token_accuracy": 0.19238465130329133,
|
|
"num_tokens": 7574149.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.509900856018066,
|
|
"epoch": 3.328941390277147,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042569441308138204,
|
|
"loss": 5.1101,
|
|
"mean_token_accuracy": 0.1972552001476288,
|
|
"num_tokens": 7584610.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.532596063613892,
|
|
"epoch": 3.3334847796456155,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004254317764298645,
|
|
"loss": 5.0685,
|
|
"mean_token_accuracy": 0.19591612219810486,
|
|
"num_tokens": 7594954.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.606906747817993,
|
|
"epoch": 3.3380281690140845,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004251687686028927,
|
|
"loss": 5.188,
|
|
"mean_token_accuracy": 0.17925947308540344,
|
|
"num_tokens": 7606069.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.477873706817627,
|
|
"epoch": 3.3425715583825535,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042490539024941216,
|
|
"loss": 5.0263,
|
|
"mean_token_accuracy": 0.1998884215950966,
|
|
"num_tokens": 7615671.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.542865562438965,
|
|
"epoch": 3.347114947751022,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042464164201928286,
|
|
"loss": 5.2178,
|
|
"mean_token_accuracy": 0.18188869059085847,
|
|
"num_tokens": 7626740.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.584362316131592,
|
|
"epoch": 3.351658337119491,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00042437752456327734,
|
|
"loss": 5.1361,
|
|
"mean_token_accuracy": 0.19445003271102906,
|
|
"num_tokens": 7636459.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.590976190567017,
|
|
"epoch": 3.35620172648796,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042411303853307907,
|
|
"loss": 5.1022,
|
|
"mean_token_accuracy": 0.19548312872648238,
|
|
"num_tokens": 7646974.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.488025379180908,
|
|
"epoch": 3.360745115856429,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004238481845812811,
|
|
"loss": 5.0612,
|
|
"mean_token_accuracy": 0.1968844309449196,
|
|
"num_tokens": 7657372.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.478169918060303,
|
|
"epoch": 3.365288505224898,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00042358296336138436,
|
|
"loss": 5.1141,
|
|
"mean_token_accuracy": 0.19376225620508195,
|
|
"num_tokens": 7668972.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.593365430831909,
|
|
"epoch": 3.3698318945933665,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004233173755277956,
|
|
"loss": 5.1243,
|
|
"mean_token_accuracy": 0.20784278512001036,
|
|
"num_tokens": 7679742.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.555311059951782,
|
|
"epoch": 3.3743752839618355,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042305142173582657,
|
|
"loss": 5.1501,
|
|
"mean_token_accuracy": 0.19712294340133668,
|
|
"num_tokens": 7690759.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.530154848098755,
|
|
"epoch": 3.3789186733303045,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00042278510264169165,
|
|
"loss": 5.1248,
|
|
"mean_token_accuracy": 0.19189250767230986,
|
|
"num_tokens": 7700482.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.500243616104126,
|
|
"epoch": 3.3834620626987735,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00042251841890250696,
|
|
"loss": 5.0863,
|
|
"mean_token_accuracy": 0.18868559300899507,
|
|
"num_tokens": 7710414.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.582386922836304,
|
|
"epoch": 3.388005452067242,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000422251371176288,
|
|
"loss": 5.147,
|
|
"mean_token_accuracy": 0.19153689444065095,
|
|
"num_tokens": 7721214.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.517453145980835,
|
|
"epoch": 3.392548841435711,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00042198396012194854,
|
|
"loss": 5.0942,
|
|
"mean_token_accuracy": 0.1952054128050804,
|
|
"num_tokens": 7732551.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.537407350540161,
|
|
"epoch": 3.39709223080418,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042171618639929865,
|
|
"loss": 5.1929,
|
|
"mean_token_accuracy": 0.1911899447441101,
|
|
"num_tokens": 7742266.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.517155456542969,
|
|
"epoch": 3.401635620172649,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00042144805066904345,
|
|
"loss": 5.0287,
|
|
"mean_token_accuracy": 0.19092843532562256,
|
|
"num_tokens": 7753152.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.496455287933349,
|
|
"epoch": 3.4061790095411175,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00042117955359278125,
|
|
"loss": 5.1438,
|
|
"mean_token_accuracy": 0.18971190899610518,
|
|
"num_tokens": 7764420.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.549328470230103,
|
|
"epoch": 3.4107223989095865,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042091069583300177,
|
|
"loss": 5.1144,
|
|
"mean_token_accuracy": 0.19185424149036406,
|
|
"num_tokens": 7776044.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.511335134506226,
|
|
"epoch": 3.4152657882780555,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004206414780530849,
|
|
"loss": 5.1432,
|
|
"mean_token_accuracy": 0.19023818373680115,
|
|
"num_tokens": 7786737.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.507075214385987,
|
|
"epoch": 3.4198091776465245,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004203719009172987,
|
|
"loss": 5.0621,
|
|
"mean_token_accuracy": 0.1944119945168495,
|
|
"num_tokens": 7796974.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.560242462158203,
|
|
"epoch": 3.424352567014993,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000420101965090798,
|
|
"loss": 5.2332,
|
|
"mean_token_accuracy": 0.18790064007043839,
|
|
"num_tokens": 7808221.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.495210313796997,
|
|
"epoch": 3.428895956383462,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004198316712396226,
|
|
"loss": 5.069,
|
|
"mean_token_accuracy": 0.19481909424066543,
|
|
"num_tokens": 7818422.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.5231067657470705,
|
|
"epoch": 3.433439345751931,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00041956102003069573,
|
|
"loss": 5.0895,
|
|
"mean_token_accuracy": 0.19694193750619887,
|
|
"num_tokens": 7828313.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.4698442935943605,
|
|
"epoch": 3.4379827351204,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004192900121318223,
|
|
"loss": 5.1165,
|
|
"mean_token_accuracy": 0.2042548418045044,
|
|
"num_tokens": 7839493.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.527283954620361,
|
|
"epoch": 3.4425261244888685,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004190186482116874,
|
|
"loss": 5.1172,
|
|
"mean_token_accuracy": 0.18893566727638245,
|
|
"num_tokens": 7849796.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.5383843898773195,
|
|
"epoch": 3.4470695138573375,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004187469289398545,
|
|
"loss": 5.1649,
|
|
"mean_token_accuracy": 0.1902869462966919,
|
|
"num_tokens": 7860865.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.491477108001709,
|
|
"epoch": 3.4516129032258065,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041847485498676385,
|
|
"loss": 5.102,
|
|
"mean_token_accuracy": 0.20021062195301056,
|
|
"num_tokens": 7871144.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.487704801559448,
|
|
"epoch": 3.4561562925942755,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00041820242702373094,
|
|
"loss": 5.1482,
|
|
"mean_token_accuracy": 0.19539811164140702,
|
|
"num_tokens": 7880916.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.467694282531738,
|
|
"epoch": 3.460699681962744,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004179296457229448,
|
|
"loss": 5.0869,
|
|
"mean_token_accuracy": 0.19446915686130523,
|
|
"num_tokens": 7891813.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.548619985580444,
|
|
"epoch": 3.465243071331213,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004176565117574659,
|
|
"loss": 5.1485,
|
|
"mean_token_accuracy": 0.1929413765668869,
|
|
"num_tokens": 7902697.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.516193580627442,
|
|
"epoch": 3.469786460699682,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004173830258012255,
|
|
"loss": 5.1193,
|
|
"mean_token_accuracy": 0.1948766142129898,
|
|
"num_tokens": 7912777.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.533999013900757,
|
|
"epoch": 3.474329850068151,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004171091885290226,
|
|
"loss": 5.1653,
|
|
"mean_token_accuracy": 0.19172380566596986,
|
|
"num_tokens": 7923255.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.5096688747406,
|
|
"epoch": 3.4788732394366195,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004168350006165238,
|
|
"loss": 5.0372,
|
|
"mean_token_accuracy": 0.1923026219010353,
|
|
"num_tokens": 7934268.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.440487384796143,
|
|
"epoch": 3.4834166288050885,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041656046274026034,
|
|
"loss": 5.0546,
|
|
"mean_token_accuracy": 0.19613176584243774,
|
|
"num_tokens": 7944526.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.527753400802612,
|
|
"epoch": 3.4879600181735575,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004162855755776272,
|
|
"loss": 5.1583,
|
|
"mean_token_accuracy": 0.1898191452026367,
|
|
"num_tokens": 7954512.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.516797399520874,
|
|
"epoch": 3.4925034075420265,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00041601033980688106,
|
|
"loss": 5.175,
|
|
"mean_token_accuracy": 0.18098063468933107,
|
|
"num_tokens": 7965703.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.56067042350769,
|
|
"epoch": 3.497046796910495,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004157347561071389,
|
|
"loss": 5.154,
|
|
"mean_token_accuracy": 0.19192345440387726,
|
|
"num_tokens": 7977822.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.491831111907959,
|
|
"epoch": 3.501590186278964,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00041545882515837604,
|
|
"loss": 5.0652,
|
|
"mean_token_accuracy": 0.1951443776488304,
|
|
"num_tokens": 7988318.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.521320056915283,
|
|
"epoch": 3.506133575647433,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00041518254764142477,
|
|
"loss": 5.039,
|
|
"mean_token_accuracy": 0.19523254483938218,
|
|
"num_tokens": 7998076.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.495612287521363,
|
|
"epoch": 3.510676965015902,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00041490592423797226,
|
|
"loss": 5.1417,
|
|
"mean_token_accuracy": 0.1977503314614296,
|
|
"num_tokens": 8007913.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.500504207611084,
|
|
"epoch": 3.515220354384371,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004146289556305594,
|
|
"loss": 5.1017,
|
|
"mean_token_accuracy": 0.19287894964218139,
|
|
"num_tokens": 8018340.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.579501390457153,
|
|
"epoch": 3.5197637437528395,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004143516425025786,
|
|
"loss": 5.1546,
|
|
"mean_token_accuracy": 0.19088564664125443,
|
|
"num_tokens": 8028312.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.602171230316162,
|
|
"epoch": 3.5243071331213085,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004140739855382724,
|
|
"loss": 5.1802,
|
|
"mean_token_accuracy": 0.1930556297302246,
|
|
"num_tokens": 8038580.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.572351694107056,
|
|
"epoch": 3.5288505224897775,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041379598542273197,
|
|
"loss": 5.1241,
|
|
"mean_token_accuracy": 0.1891513153910637,
|
|
"num_tokens": 8048365.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.500466442108154,
|
|
"epoch": 3.533393911858246,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00041351764284189464,
|
|
"loss": 5.1861,
|
|
"mean_token_accuracy": 0.18407674580812455,
|
|
"num_tokens": 8059638.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.588624143600464,
|
|
"epoch": 3.537937301226715,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004132389584825433,
|
|
"loss": 5.2521,
|
|
"mean_token_accuracy": 0.1884332224726677,
|
|
"num_tokens": 8070993.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.517414808273315,
|
|
"epoch": 3.542480690595184,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041295993303230383,
|
|
"loss": 5.0695,
|
|
"mean_token_accuracy": 0.2009451150894165,
|
|
"num_tokens": 8081203.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.4289271354675295,
|
|
"epoch": 3.547024079963653,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004126805671796438,
|
|
"loss": 5.09,
|
|
"mean_token_accuracy": 0.19282526522874832,
|
|
"num_tokens": 8091564.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.501790571212768,
|
|
"epoch": 3.551567469332122,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004124008616138707,
|
|
"loss": 5.0655,
|
|
"mean_token_accuracy": 0.1935689076781273,
|
|
"num_tokens": 8100675.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.493474006652832,
|
|
"epoch": 3.5561108587005905,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00041212081702513015,
|
|
"loss": 5.1311,
|
|
"mean_token_accuracy": 0.1991216644644737,
|
|
"num_tokens": 8110605.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.517332744598389,
|
|
"epoch": 3.5606542480690595,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004118404341044044,
|
|
"loss": 5.1728,
|
|
"mean_token_accuracy": 0.18816904127597808,
|
|
"num_tokens": 8120910.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.550918817520142,
|
|
"epoch": 3.5651976374375285,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00041155971354351034,
|
|
"loss": 5.1224,
|
|
"mean_token_accuracy": 0.19645454734563828,
|
|
"num_tokens": 8131827.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.515909624099732,
|
|
"epoch": 3.569741026805997,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00041127865603509815,
|
|
"loss": 5.1014,
|
|
"mean_token_accuracy": 0.19348724335432052,
|
|
"num_tokens": 8141785.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.567871189117431,
|
|
"epoch": 3.574284416174466,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004109972622726493,
|
|
"loss": 5.1171,
|
|
"mean_token_accuracy": 0.18949676901102067,
|
|
"num_tokens": 8151732.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.519099378585816,
|
|
"epoch": 3.578827805542935,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00041071553295047485,
|
|
"loss": 5.2124,
|
|
"mean_token_accuracy": 0.18984898924827576,
|
|
"num_tokens": 8162334.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.494318151473999,
|
|
"epoch": 3.583371194911404,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004104334687637139,
|
|
"loss": 5.0898,
|
|
"mean_token_accuracy": 0.19082968831062316,
|
|
"num_tokens": 8172810.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.487064933776855,
|
|
"epoch": 3.587914584279873,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00041015107040833187,
|
|
"loss": 5.123,
|
|
"mean_token_accuracy": 0.19004584550857545,
|
|
"num_tokens": 8183240.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.539603424072266,
|
|
"epoch": 3.5924579736483415,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004098683385811185,
|
|
"loss": 5.0684,
|
|
"mean_token_accuracy": 0.19060366749763488,
|
|
"num_tokens": 8193824.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.501092910766602,
|
|
"epoch": 3.5970013630168105,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00040958527397968666,
|
|
"loss": 5.0606,
|
|
"mean_token_accuracy": 0.2012089416384697,
|
|
"num_tokens": 8203792.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.539542484283447,
|
|
"epoch": 3.6015447523852795,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004093018773024699,
|
|
"loss": 5.2436,
|
|
"mean_token_accuracy": 0.18750120401382447,
|
|
"num_tokens": 8214436.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.591252279281616,
|
|
"epoch": 3.606088141753748,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004090181492487215,
|
|
"loss": 5.1024,
|
|
"mean_token_accuracy": 0.19021122008562089,
|
|
"num_tokens": 8224518.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.5576845645904545,
|
|
"epoch": 3.610631531122217,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00040873409051851223,
|
|
"loss": 5.2597,
|
|
"mean_token_accuracy": 0.18063579946756364,
|
|
"num_tokens": 8235234.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.454524946212769,
|
|
"epoch": 3.615174920490686,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004084497018127286,
|
|
"loss": 5.1325,
|
|
"mean_token_accuracy": 0.1922066777944565,
|
|
"num_tokens": 8246612.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.57022647857666,
|
|
"epoch": 3.619718309859155,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004081649838330717,
|
|
"loss": 5.0924,
|
|
"mean_token_accuracy": 0.19999369531869887,
|
|
"num_tokens": 8256720.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.527831506729126,
|
|
"epoch": 3.624261699227624,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00040787993728205464,
|
|
"loss": 5.1514,
|
|
"mean_token_accuracy": 0.1955747663974762,
|
|
"num_tokens": 8267666.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.563654661178589,
|
|
"epoch": 3.6288050885960925,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00040759456286300154,
|
|
"loss": 5.137,
|
|
"mean_token_accuracy": 0.19323565512895585,
|
|
"num_tokens": 8277882.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.510303163528443,
|
|
"epoch": 3.6333484779645615,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00040730886128004544,
|
|
"loss": 5.0526,
|
|
"mean_token_accuracy": 0.19388376027345658,
|
|
"num_tokens": 8286968.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.6333484779645615,
|
|
"eval_entropy": 5.339761698523233,
|
|
"eval_loss": 5.732693195343018,
|
|
"eval_mean_token_accuracy": 0.1681636985478013,
|
|
"eval_num_tokens": 8286968.0,
|
|
"eval_runtime": 2.0081,
|
|
"eval_samples_per_second": 1710.543,
|
|
"eval_steps_per_second": 214.129,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.48574538230896,
|
|
"epoch": 3.6378918673330305,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004070228332381266,
|
|
"loss": 5.124,
|
|
"mean_token_accuracy": 0.19649865329265595,
|
|
"num_tokens": 8297154.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.528165864944458,
|
|
"epoch": 3.6424352567014995,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004067364794429908,
|
|
"loss": 5.1153,
|
|
"mean_token_accuracy": 0.18957486748695374,
|
|
"num_tokens": 8307954.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.439371633529663,
|
|
"epoch": 3.646978646069968,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004064498006011874,
|
|
"loss": 5.1032,
|
|
"mean_token_accuracy": 0.19430603235960006,
|
|
"num_tokens": 8317714.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.455359363555909,
|
|
"epoch": 3.651522035438437,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00040616279742006813,
|
|
"loss": 5.1272,
|
|
"mean_token_accuracy": 0.19905939996242522,
|
|
"num_tokens": 8327963.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.518960332870483,
|
|
"epoch": 3.656065424806906,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040587547060778466,
|
|
"loss": 5.0536,
|
|
"mean_token_accuracy": 0.19981419891119004,
|
|
"num_tokens": 8338190.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.4930586338043215,
|
|
"epoch": 3.660608814175375,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004055878208732874,
|
|
"loss": 5.1867,
|
|
"mean_token_accuracy": 0.18996217846870422,
|
|
"num_tokens": 8348564.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.491205930709839,
|
|
"epoch": 3.665152203543844,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040529984892632363,
|
|
"loss": 5.1108,
|
|
"mean_token_accuracy": 0.1929684728384018,
|
|
"num_tokens": 8359463.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.534110736846924,
|
|
"epoch": 3.6696955929123125,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004050115554774353,
|
|
"loss": 5.161,
|
|
"mean_token_accuracy": 0.1902874156832695,
|
|
"num_tokens": 8369676.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.5096046924591064,
|
|
"epoch": 3.6742389822807815,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004047229412379578,
|
|
"loss": 5.0685,
|
|
"mean_token_accuracy": 0.20735878646373748,
|
|
"num_tokens": 8379579.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.458685493469238,
|
|
"epoch": 3.6787823716492505,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004044340069200183,
|
|
"loss": 5.1359,
|
|
"mean_token_accuracy": 0.1940714970231056,
|
|
"num_tokens": 8390164.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.45562710762024,
|
|
"epoch": 3.683325761017719,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00040414475323653326,
|
|
"loss": 5.0379,
|
|
"mean_token_accuracy": 0.19735147804021835,
|
|
"num_tokens": 8399220.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.522087240219117,
|
|
"epoch": 3.687869150386188,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004038551809012076,
|
|
"loss": 5.1109,
|
|
"mean_token_accuracy": 0.1911008045077324,
|
|
"num_tokens": 8409488.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.511661338806152,
|
|
"epoch": 3.692412539754657,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004035652906285322,
|
|
"loss": 5.0646,
|
|
"mean_token_accuracy": 0.20466965436935425,
|
|
"num_tokens": 8418817.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.493175840377807,
|
|
"epoch": 3.696955929123126,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00040327508313378244,
|
|
"loss": 5.1845,
|
|
"mean_token_accuracy": 0.1912352204322815,
|
|
"num_tokens": 8428742.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.468400144577027,
|
|
"epoch": 3.701499318491595,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004029845591330164,
|
|
"loss": 5.0463,
|
|
"mean_token_accuracy": 0.20120988190174102,
|
|
"num_tokens": 8438016.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.527812767028808,
|
|
"epoch": 3.7060427078600635,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00040269371934307336,
|
|
"loss": 5.1578,
|
|
"mean_token_accuracy": 0.1931580826640129,
|
|
"num_tokens": 8448428.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.4827581405639645,
|
|
"epoch": 3.7105860972285325,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00040240256448157146,
|
|
"loss": 5.0084,
|
|
"mean_token_accuracy": 0.20443811118602753,
|
|
"num_tokens": 8458444.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.471077108383179,
|
|
"epoch": 3.7151294865970015,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004021110952669065,
|
|
"loss": 5.1563,
|
|
"mean_token_accuracy": 0.19028250724077225,
|
|
"num_tokens": 8467721.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.425581216812134,
|
|
"epoch": 3.71967287596547,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004018193124182497,
|
|
"loss": 5.0716,
|
|
"mean_token_accuracy": 0.19859496504068375,
|
|
"num_tokens": 8477484.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.472807693481445,
|
|
"epoch": 3.724216265333939,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004015272166555462,
|
|
"loss": 5.0762,
|
|
"mean_token_accuracy": 0.19369181841611863,
|
|
"num_tokens": 8486902.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.495205402374268,
|
|
"epoch": 3.728759654702408,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004012348086995132,
|
|
"loss": 5.053,
|
|
"mean_token_accuracy": 0.19411021918058396,
|
|
"num_tokens": 8497626.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.563983583450318,
|
|
"epoch": 3.733303044070877,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004009420892716385,
|
|
"loss": 5.1534,
|
|
"mean_token_accuracy": 0.18233701437711716,
|
|
"num_tokens": 8507518.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.461269235610962,
|
|
"epoch": 3.737846433439346,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004006490590941781,
|
|
"loss": 5.1118,
|
|
"mean_token_accuracy": 0.19401460140943527,
|
|
"num_tokens": 8517835.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.506267118453979,
|
|
"epoch": 3.7423898228078145,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004003557188901547,
|
|
"loss": 5.167,
|
|
"mean_token_accuracy": 0.19231697171926498,
|
|
"num_tokens": 8527552.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.538556289672852,
|
|
"epoch": 3.7469332121762835,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004000620693833562,
|
|
"loss": 5.1421,
|
|
"mean_token_accuracy": 0.19452222883701326,
|
|
"num_tokens": 8538082.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.461449003219604,
|
|
"epoch": 3.7514766015447525,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003997681112983335,
|
|
"loss": 5.0794,
|
|
"mean_token_accuracy": 0.19179436266422273,
|
|
"num_tokens": 8548558.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.4641234397888185,
|
|
"epoch": 3.756019990913221,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039947384536039905,
|
|
"loss": 5.1063,
|
|
"mean_token_accuracy": 0.1959160566329956,
|
|
"num_tokens": 8559406.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.5509367942810055,
|
|
"epoch": 3.76056338028169,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039917927229562474,
|
|
"loss": 5.1229,
|
|
"mean_token_accuracy": 0.1882578909397125,
|
|
"num_tokens": 8569519.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.449975681304932,
|
|
"epoch": 3.765106769650159,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00039888439283084035,
|
|
"loss": 5.137,
|
|
"mean_token_accuracy": 0.19129497855901717,
|
|
"num_tokens": 8579658.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.473759651184082,
|
|
"epoch": 3.769650159018628,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039858920769363164,
|
|
"loss": 5.0596,
|
|
"mean_token_accuracy": 0.20096141546964646,
|
|
"num_tokens": 8589681.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.521468591690064,
|
|
"epoch": 3.774193548387097,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003982937176123386,
|
|
"loss": 5.1408,
|
|
"mean_token_accuracy": 0.192432764172554,
|
|
"num_tokens": 8600310.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.495010232925415,
|
|
"epoch": 3.7787369377555655,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039799792331605364,
|
|
"loss": 5.1435,
|
|
"mean_token_accuracy": 0.19155607968568802,
|
|
"num_tokens": 8610920.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.423186635971069,
|
|
"epoch": 3.7832803271240345,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00039770182553461965,
|
|
"loss": 5.0735,
|
|
"mean_token_accuracy": 0.19816265851259232,
|
|
"num_tokens": 8620536.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.492016553878784,
|
|
"epoch": 3.7878237164925035,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039740542499862866,
|
|
"loss": 5.1616,
|
|
"mean_token_accuracy": 0.19480880200862885,
|
|
"num_tokens": 8631647.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.5218194961547855,
|
|
"epoch": 3.792367105860972,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003971087224394194,
|
|
"loss": 5.1237,
|
|
"mean_token_accuracy": 0.18958858251571656,
|
|
"num_tokens": 8642659.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.494124937057495,
|
|
"epoch": 3.796910495229441,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003968117185890761,
|
|
"loss": 5.1237,
|
|
"mean_token_accuracy": 0.19752606749534607,
|
|
"num_tokens": 8651775.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.497392082214356,
|
|
"epoch": 3.80145388459791,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039651441418042606,
|
|
"loss": 5.1016,
|
|
"mean_token_accuracy": 0.18746509850025178,
|
|
"num_tokens": 8662520.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.516363048553467,
|
|
"epoch": 3.805997273966379,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003962168099470384,
|
|
"loss": 5.1179,
|
|
"mean_token_accuracy": 0.19674220085144042,
|
|
"num_tokens": 8672630.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.5318756103515625,
|
|
"epoch": 3.810540663334848,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000395918906623222,
|
|
"loss": 5.1927,
|
|
"mean_token_accuracy": 0.18443794101476668,
|
|
"num_tokens": 8683348.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.450568151473999,
|
|
"epoch": 3.8150840527033165,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003956207049440238,
|
|
"loss": 5.0487,
|
|
"mean_token_accuracy": 0.19580661058425902,
|
|
"num_tokens": 8693423.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.526213073730469,
|
|
"epoch": 3.8196274420717855,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003953222056452265,
|
|
"loss": 5.1251,
|
|
"mean_token_accuracy": 0.19483501315116883,
|
|
"num_tokens": 8703766.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.5830193042755125,
|
|
"epoch": 3.8241708314402545,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003950234094633479,
|
|
"loss": 5.2132,
|
|
"mean_token_accuracy": 0.189988574385643,
|
|
"num_tokens": 8715330.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.486254930496216,
|
|
"epoch": 3.828714220808723,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039472431713563746,
|
|
"loss": 5.0916,
|
|
"mean_token_accuracy": 0.19512208551168442,
|
|
"num_tokens": 8725732.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.580050992965698,
|
|
"epoch": 3.833257610177192,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000394424929400076,
|
|
"loss": 5.0827,
|
|
"mean_token_accuracy": 0.20025289803743362,
|
|
"num_tokens": 8736377.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.551922225952149,
|
|
"epoch": 3.837800999545661,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003941252469953729,
|
|
"loss": 5.1134,
|
|
"mean_token_accuracy": 0.19279252290725707,
|
|
"num_tokens": 8746991.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.427538585662842,
|
|
"epoch": 3.84234438891413,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003938252706609647,
|
|
"loss": 5.1186,
|
|
"mean_token_accuracy": 0.19493389427661895,
|
|
"num_tokens": 8756920.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.503057765960693,
|
|
"epoch": 3.846887778282599,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003935250011370132,
|
|
"loss": 5.1853,
|
|
"mean_token_accuracy": 0.1817367836833,
|
|
"num_tokens": 8768617.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.56388521194458,
|
|
"epoch": 3.851431167651068,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003932244391644037,
|
|
"loss": 5.1166,
|
|
"mean_token_accuracy": 0.204342257976532,
|
|
"num_tokens": 8778427.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.486303043365479,
|
|
"epoch": 3.8559745570195365,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003929235854847428,
|
|
"loss": 5.1555,
|
|
"mean_token_accuracy": 0.19356151968240737,
|
|
"num_tokens": 8789093.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.491094779968262,
|
|
"epoch": 3.8605179463880055,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039262244084035724,
|
|
"loss": 5.1009,
|
|
"mean_token_accuracy": 0.189717660844326,
|
|
"num_tokens": 8799832.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.527243328094483,
|
|
"epoch": 3.8650613357564745,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00039232100597429143,
|
|
"loss": 5.131,
|
|
"mean_token_accuracy": 0.19439983516931533,
|
|
"num_tokens": 8810149.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.527439880371094,
|
|
"epoch": 3.869604725124943,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00039201928163030584,
|
|
"loss": 5.1367,
|
|
"mean_token_accuracy": 0.1978183090686798,
|
|
"num_tokens": 8820087.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.4579802513122555,
|
|
"epoch": 3.874148114493412,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00039171726855287544,
|
|
"loss": 5.1962,
|
|
"mean_token_accuracy": 0.18271831423044205,
|
|
"num_tokens": 8830959.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.550218486785889,
|
|
"epoch": 3.878691503861881,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00039141496748718735,
|
|
"loss": 5.1291,
|
|
"mean_token_accuracy": 0.19162626564502716,
|
|
"num_tokens": 8842522.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.443166780471802,
|
|
"epoch": 3.88323489323035,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003911123791791395,
|
|
"loss": 5.0376,
|
|
"mean_token_accuracy": 0.19930487275123596,
|
|
"num_tokens": 8852521.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.585235452651977,
|
|
"epoch": 3.887778282598819,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003908095043753384,
|
|
"loss": 5.1704,
|
|
"mean_token_accuracy": 0.18953115344047547,
|
|
"num_tokens": 8862156.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.543653392791748,
|
|
"epoch": 3.8923216719672875,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039050634382309753,
|
|
"loss": 5.0392,
|
|
"mean_token_accuracy": 0.19603938460350037,
|
|
"num_tokens": 8872772.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.483927822113037,
|
|
"epoch": 3.8968650613357565,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003902028982704355,
|
|
"loss": 5.1267,
|
|
"mean_token_accuracy": 0.196218042075634,
|
|
"num_tokens": 8883229.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.4501832008361815,
|
|
"epoch": 3.9014084507042255,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003898991684660738,
|
|
"loss": 5.1101,
|
|
"mean_token_accuracy": 0.1930313304066658,
|
|
"num_tokens": 8892403.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.481894302368164,
|
|
"epoch": 3.905951840072694,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003895951551594358,
|
|
"loss": 5.1314,
|
|
"mean_token_accuracy": 0.18622865676879882,
|
|
"num_tokens": 8902617.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.575549507141114,
|
|
"epoch": 3.910495229441163,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000389290859100644,
|
|
"loss": 5.1614,
|
|
"mean_token_accuracy": 0.19435564577579498,
|
|
"num_tokens": 8913099.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.438909721374512,
|
|
"epoch": 3.915038618809632,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003889862810405187,
|
|
"loss": 5.0622,
|
|
"mean_token_accuracy": 0.1944717675447464,
|
|
"num_tokens": 8922469.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.487117195129395,
|
|
"epoch": 3.919582008178101,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000388681421730576,
|
|
"loss": 5.1475,
|
|
"mean_token_accuracy": 0.19636933207511903,
|
|
"num_tokens": 8933261.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.516995763778686,
|
|
"epoch": 3.92412539754657,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003883762819230259,
|
|
"loss": 5.1325,
|
|
"mean_token_accuracy": 0.19926582723855973,
|
|
"num_tokens": 8943873.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.489044666290283,
|
|
"epoch": 3.9286687869150385,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00038807086237077057,
|
|
"loss": 5.0417,
|
|
"mean_token_accuracy": 0.2082255631685257,
|
|
"num_tokens": 8953802.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.550192213058471,
|
|
"epoch": 3.9332121762835075,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038776516382740237,
|
|
"loss": 5.2359,
|
|
"mean_token_accuracy": 0.18553104549646376,
|
|
"num_tokens": 8964309.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.50753755569458,
|
|
"epoch": 3.9377555656519765,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00038745918704720205,
|
|
"loss": 5.0961,
|
|
"mean_token_accuracy": 0.1957569494843483,
|
|
"num_tokens": 8973611.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.457180738449097,
|
|
"epoch": 3.942298955020445,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00038715293278513693,
|
|
"loss": 5.1391,
|
|
"mean_token_accuracy": 0.18416973054409028,
|
|
"num_tokens": 8985783.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.523426342010498,
|
|
"epoch": 3.946842344388914,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00038684640179685896,
|
|
"loss": 5.1695,
|
|
"mean_token_accuracy": 0.18669908940792085,
|
|
"num_tokens": 8996736.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.559651231765747,
|
|
"epoch": 3.951385733757383,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003865395948387029,
|
|
"loss": 5.2281,
|
|
"mean_token_accuracy": 0.1840924322605133,
|
|
"num_tokens": 9007635.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.533151578903198,
|
|
"epoch": 3.955929123125852,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00038623251266768437,
|
|
"loss": 5.0851,
|
|
"mean_token_accuracy": 0.19194516390562058,
|
|
"num_tokens": 9017892.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.486858272552491,
|
|
"epoch": 3.960472512494321,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038592515604149815,
|
|
"loss": 5.1282,
|
|
"mean_token_accuracy": 0.1944011226296425,
|
|
"num_tokens": 9027594.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.497931671142578,
|
|
"epoch": 3.9650159018627895,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003856175257185162,
|
|
"loss": 5.1519,
|
|
"mean_token_accuracy": 0.1919327273964882,
|
|
"num_tokens": 9037083.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.529991817474365,
|
|
"epoch": 3.9695592912312585,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038530962245778564,
|
|
"loss": 5.1078,
|
|
"mean_token_accuracy": 0.2003428414463997,
|
|
"num_tokens": 9047176.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.527696943283081,
|
|
"epoch": 3.9741026805997275,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003850014470190272,
|
|
"loss": 5.0773,
|
|
"mean_token_accuracy": 0.19598610252141951,
|
|
"num_tokens": 9057810.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.5210377216339115,
|
|
"epoch": 3.978646069968196,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003846930001626332,
|
|
"loss": 5.1748,
|
|
"mean_token_accuracy": 0.19256748706102372,
|
|
"num_tokens": 9067765.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.559467077255249,
|
|
"epoch": 3.983189459336665,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003843842826496656,
|
|
"loss": 5.1242,
|
|
"mean_token_accuracy": 0.19140773862600327,
|
|
"num_tokens": 9078796.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.495080089569091,
|
|
"epoch": 3.987732848705134,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000384075295241854,
|
|
"loss": 5.0906,
|
|
"mean_token_accuracy": 0.19998400807380676,
|
|
"num_tokens": 9088368.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.444483232498169,
|
|
"epoch": 3.992276238073603,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003837660387015944,
|
|
"loss": 5.0332,
|
|
"mean_token_accuracy": 0.1946394845843315,
|
|
"num_tokens": 9099082.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.506217098236084,
|
|
"epoch": 3.996819627442072,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003834565137919463,
|
|
"loss": 5.1626,
|
|
"mean_token_accuracy": 0.19718001782894135,
|
|
"num_tokens": 9110543.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.506849500868055,
|
|
"epoch": 4.000908677873694,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003831467212766318,
|
|
"loss": 5.1077,
|
|
"mean_token_accuracy": 0.19700109958648682,
|
|
"num_tokens": 9119347.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.444593238830566,
|
|
"epoch": 4.005452067242163,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003828366619200331,
|
|
"loss": 4.7887,
|
|
"mean_token_accuracy": 0.21329672485589982,
|
|
"num_tokens": 9129154.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.408209991455078,
|
|
"epoch": 4.009995456610632,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003825263364871908,
|
|
"loss": 4.8268,
|
|
"mean_token_accuracy": 0.21090005189180375,
|
|
"num_tokens": 9139877.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.461040878295899,
|
|
"epoch": 4.0145388459791,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003822157457438022,
|
|
"loss": 4.8806,
|
|
"mean_token_accuracy": 0.21466604471206666,
|
|
"num_tokens": 9150508.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.458488464355469,
|
|
"epoch": 4.019082235347569,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038190489045621893,
|
|
"loss": 4.8318,
|
|
"mean_token_accuracy": 0.21233197450637817,
|
|
"num_tokens": 9160403.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.57012038230896,
|
|
"epoch": 4.023625624716038,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003815937713914455,
|
|
"loss": 4.8877,
|
|
"mean_token_accuracy": 0.21328169852495193,
|
|
"num_tokens": 9171026.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.458043622970581,
|
|
"epoch": 4.028169014084507,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00038128238931713743,
|
|
"loss": 4.851,
|
|
"mean_token_accuracy": 0.21564830988645553,
|
|
"num_tokens": 9181252.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.532674407958984,
|
|
"epoch": 4.032712403452976,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038097074500159883,
|
|
"loss": 4.9186,
|
|
"mean_token_accuracy": 0.20425556749105453,
|
|
"num_tokens": 9190803.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.500670146942139,
|
|
"epoch": 4.037255792821445,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038065883921378123,
|
|
"loss": 4.8957,
|
|
"mean_token_accuracy": 0.20812420547008514,
|
|
"num_tokens": 9201077.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.456028175354004,
|
|
"epoch": 4.041799182189914,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003803466727232811,
|
|
"loss": 4.918,
|
|
"mean_token_accuracy": 0.21294918954372405,
|
|
"num_tokens": 9211239.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.470507287979126,
|
|
"epoch": 4.046342571558383,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003800342463003383,
|
|
"loss": 4.8798,
|
|
"mean_token_accuracy": 0.21690812557935715,
|
|
"num_tokens": 9222683.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.472645473480225,
|
|
"epoch": 4.050885960926851,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00037972156071583386,
|
|
"loss": 4.8481,
|
|
"mean_token_accuracy": 0.21180264353752137,
|
|
"num_tokens": 9232099.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.435933303833008,
|
|
"epoch": 4.05542935029532,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003794086167412884,
|
|
"loss": 4.8953,
|
|
"mean_token_accuracy": 0.20841824412345886,
|
|
"num_tokens": 9242660.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.440661144256592,
|
|
"epoch": 4.059972739663789,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003790954151488601,
|
|
"loss": 4.9265,
|
|
"mean_token_accuracy": 0.20264368057250975,
|
|
"num_tokens": 9253981.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.485601425170898,
|
|
"epoch": 4.064516129032258,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003787819567113427,
|
|
"loss": 4.849,
|
|
"mean_token_accuracy": 0.21507286578416823,
|
|
"num_tokens": 9264262.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.397464036941528,
|
|
"epoch": 4.069059518400727,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003784682422021638,
|
|
"loss": 4.859,
|
|
"mean_token_accuracy": 0.21353696584701537,
|
|
"num_tokens": 9273562.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.47631344795227,
|
|
"epoch": 4.073602907769196,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003781542723953828,
|
|
"loss": 4.8828,
|
|
"mean_token_accuracy": 0.20807297825813292,
|
|
"num_tokens": 9283171.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.463612794876099,
|
|
"epoch": 4.078146297137665,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00037784004806568885,
|
|
"loss": 4.8618,
|
|
"mean_token_accuracy": 0.21347161531448364,
|
|
"num_tokens": 9292624.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.484060478210449,
|
|
"epoch": 4.082689686506134,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00037752556998839934,
|
|
"loss": 4.9434,
|
|
"mean_token_accuracy": 0.20481917560100554,
|
|
"num_tokens": 9303159.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.480402278900146,
|
|
"epoch": 4.087233075874602,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003772108389394576,
|
|
"loss": 4.8844,
|
|
"mean_token_accuracy": 0.2132931172847748,
|
|
"num_tokens": 9314340.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 4.087233075874602,
|
|
"eval_entropy": 5.286269603773605,
|
|
"eval_loss": 5.721370697021484,
|
|
"eval_mean_token_accuracy": 0.1693312707855258,
|
|
"eval_num_tokens": 9314340.0,
|
|
"eval_runtime": 2.0045,
|
|
"eval_samples_per_second": 1713.655,
|
|
"eval_steps_per_second": 214.519,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.509492111206055,
|
|
"epoch": 4.091776465243071,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00037689585569543137,
|
|
"loss": 4.9623,
|
|
"mean_token_accuracy": 0.21226249933242797,
|
|
"num_tokens": 9325381.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.4603965282440186,
|
|
"epoch": 4.09631985461154,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003765806210335103,
|
|
"loss": 4.9246,
|
|
"mean_token_accuracy": 0.20106051117181778,
|
|
"num_tokens": 9336639.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.4720457077026365,
|
|
"epoch": 4.100863243980009,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00037626513573150473,
|
|
"loss": 4.8394,
|
|
"mean_token_accuracy": 0.21663217842578888,
|
|
"num_tokens": 9346794.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.377048969268799,
|
|
"epoch": 4.105406633348478,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00037594940056784315,
|
|
"loss": 4.8987,
|
|
"mean_token_accuracy": 0.21495754420757293,
|
|
"num_tokens": 9356326.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.492370223999023,
|
|
"epoch": 4.109950022716947,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003756334163215708,
|
|
"loss": 4.8876,
|
|
"mean_token_accuracy": 0.2137880280613899,
|
|
"num_tokens": 9366402.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.425258588790894,
|
|
"epoch": 4.114493412085416,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00037531718377234723,
|
|
"loss": 4.8455,
|
|
"mean_token_accuracy": 0.20882460325956345,
|
|
"num_tokens": 9377067.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.4494390964508055,
|
|
"epoch": 4.119036801453885,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003750007037004451,
|
|
"loss": 4.8998,
|
|
"mean_token_accuracy": 0.20876526087522507,
|
|
"num_tokens": 9387206.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.474677038192749,
|
|
"epoch": 4.123580190822353,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003746839768867473,
|
|
"loss": 4.8814,
|
|
"mean_token_accuracy": 0.21294330209493637,
|
|
"num_tokens": 9397195.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.410915374755859,
|
|
"epoch": 4.128123580190822,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003743670041127458,
|
|
"loss": 4.8424,
|
|
"mean_token_accuracy": 0.21010585874319077,
|
|
"num_tokens": 9407611.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.470707130432129,
|
|
"epoch": 4.132666969559291,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003740497861605395,
|
|
"loss": 4.949,
|
|
"mean_token_accuracy": 0.20490022599697114,
|
|
"num_tokens": 9417897.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.457574081420899,
|
|
"epoch": 4.13721035892776,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000373732323812832,
|
|
"loss": 4.8987,
|
|
"mean_token_accuracy": 0.19851430803537368,
|
|
"num_tokens": 9428896.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.420271730422973,
|
|
"epoch": 4.141753748296229,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00037341461785293025,
|
|
"loss": 4.8067,
|
|
"mean_token_accuracy": 0.21787624657154084,
|
|
"num_tokens": 9438854.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.354292106628418,
|
|
"epoch": 4.146297137664698,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003730966690647421,
|
|
"loss": 4.8832,
|
|
"mean_token_accuracy": 0.20495354384183884,
|
|
"num_tokens": 9447798.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.441825199127197,
|
|
"epoch": 4.150840527033167,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00037277847823277444,
|
|
"loss": 4.8446,
|
|
"mean_token_accuracy": 0.22009959518909455,
|
|
"num_tokens": 9458930.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.401585102081299,
|
|
"epoch": 4.155383916401636,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00037246004614213164,
|
|
"loss": 4.8298,
|
|
"mean_token_accuracy": 0.2126998707652092,
|
|
"num_tokens": 9469740.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.414677095413208,
|
|
"epoch": 4.159927305770104,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003721413735785132,
|
|
"loss": 4.831,
|
|
"mean_token_accuracy": 0.21663186103105544,
|
|
"num_tokens": 9479823.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.42176628112793,
|
|
"epoch": 4.164470695138573,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000371822461328212,
|
|
"loss": 4.8147,
|
|
"mean_token_accuracy": 0.21695979982614516,
|
|
"num_tokens": 9489995.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.420973062515259,
|
|
"epoch": 4.169014084507042,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00037150331017811233,
|
|
"loss": 4.879,
|
|
"mean_token_accuracy": 0.21162481158971785,
|
|
"num_tokens": 9501311.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.447546815872192,
|
|
"epoch": 4.173557473875511,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003711839209156878,
|
|
"loss": 4.9344,
|
|
"mean_token_accuracy": 0.2053820461034775,
|
|
"num_tokens": 9511901.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.483423233032227,
|
|
"epoch": 4.17810086324398,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003708642943289999,
|
|
"loss": 4.9088,
|
|
"mean_token_accuracy": 0.20365366488695144,
|
|
"num_tokens": 9522908.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.506639719009399,
|
|
"epoch": 4.182644252612449,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003705444312066953,
|
|
"loss": 4.9782,
|
|
"mean_token_accuracy": 0.20541028678417206,
|
|
"num_tokens": 9533363.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.417387533187866,
|
|
"epoch": 4.187187641980918,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003702243323380045,
|
|
"loss": 4.8445,
|
|
"mean_token_accuracy": 0.2103462114930153,
|
|
"num_tokens": 9543302.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.420156669616699,
|
|
"epoch": 4.191731031349387,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00036990399851273965,
|
|
"loss": 4.8792,
|
|
"mean_token_accuracy": 0.20139023810625076,
|
|
"num_tokens": 9553143.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.492001867294311,
|
|
"epoch": 4.196274420717856,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003695834305212925,
|
|
"loss": 4.9094,
|
|
"mean_token_accuracy": 0.20850615352392196,
|
|
"num_tokens": 9564236.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.4656729221344,
|
|
"epoch": 4.200817810086324,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00036926262915463294,
|
|
"loss": 4.8709,
|
|
"mean_token_accuracy": 0.21142637133598327,
|
|
"num_tokens": 9574363.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.38117413520813,
|
|
"epoch": 4.205361199454793,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036894159520430616,
|
|
"loss": 4.9115,
|
|
"mean_token_accuracy": 0.2039177820086479,
|
|
"num_tokens": 9585631.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.410091686248779,
|
|
"epoch": 4.209904588823262,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036862032946243173,
|
|
"loss": 4.8874,
|
|
"mean_token_accuracy": 0.20500327050685882,
|
|
"num_tokens": 9596826.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.342634820938111,
|
|
"epoch": 4.214447978191731,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00036829883272170087,
|
|
"loss": 4.8371,
|
|
"mean_token_accuracy": 0.208853779733181,
|
|
"num_tokens": 9608158.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.4350152015686035,
|
|
"epoch": 4.2189913675602,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00036797710577537486,
|
|
"loss": 4.9452,
|
|
"mean_token_accuracy": 0.20586614757776261,
|
|
"num_tokens": 9619437.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.483894491195679,
|
|
"epoch": 4.223534756928669,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036765514941728295,
|
|
"loss": 4.9328,
|
|
"mean_token_accuracy": 0.2143984854221344,
|
|
"num_tokens": 9629913.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.422917222976684,
|
|
"epoch": 4.228078146297138,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003673329644418204,
|
|
"loss": 4.8755,
|
|
"mean_token_accuracy": 0.21288909018039703,
|
|
"num_tokens": 9640033.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.4243927001953125,
|
|
"epoch": 4.232621535665607,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00036701055164394664,
|
|
"loss": 4.9389,
|
|
"mean_token_accuracy": 0.20443956106901168,
|
|
"num_tokens": 9651010.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.417495059967041,
|
|
"epoch": 4.237164925034075,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00036668791181918313,
|
|
"loss": 4.8501,
|
|
"mean_token_accuracy": 0.21399904936552047,
|
|
"num_tokens": 9661549.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.425372648239136,
|
|
"epoch": 4.241708314402544,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003663650457636117,
|
|
"loss": 4.9032,
|
|
"mean_token_accuracy": 0.20351112484931946,
|
|
"num_tokens": 9671619.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.508477783203125,
|
|
"epoch": 4.246251703771013,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003660419542738722,
|
|
"loss": 4.9452,
|
|
"mean_token_accuracy": 0.2129849061369896,
|
|
"num_tokens": 9682582.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.400639820098877,
|
|
"epoch": 4.250795093139482,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003657186381471608,
|
|
"loss": 4.8548,
|
|
"mean_token_accuracy": 0.21609467566013335,
|
|
"num_tokens": 9691997.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.434220886230468,
|
|
"epoch": 4.255338482507951,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00036539509818122765,
|
|
"loss": 4.9271,
|
|
"mean_token_accuracy": 0.20817471742630006,
|
|
"num_tokens": 9702105.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.459644222259522,
|
|
"epoch": 4.25988187187642,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003650713351743758,
|
|
"loss": 4.9091,
|
|
"mean_token_accuracy": 0.21257996708154678,
|
|
"num_tokens": 9712347.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.406892204284668,
|
|
"epoch": 4.264425261244889,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003647473499254581,
|
|
"loss": 4.9152,
|
|
"mean_token_accuracy": 0.20855187624692917,
|
|
"num_tokens": 9723116.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.389953374862671,
|
|
"epoch": 4.268968650613358,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000364423143233876,
|
|
"loss": 4.8366,
|
|
"mean_token_accuracy": 0.21201517432928085,
|
|
"num_tokens": 9732656.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.470339822769165,
|
|
"epoch": 4.273512039981826,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00036409871589957723,
|
|
"loss": 4.9587,
|
|
"mean_token_accuracy": 0.20791341215372086,
|
|
"num_tokens": 9743312.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.4355579853057865,
|
|
"epoch": 4.278055429350295,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000363774068723054,
|
|
"loss": 4.8969,
|
|
"mean_token_accuracy": 0.20924800783395767,
|
|
"num_tokens": 9753358.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.401388216018677,
|
|
"epoch": 4.282598818718764,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00036344920250534106,
|
|
"loss": 4.806,
|
|
"mean_token_accuracy": 0.21646466255187988,
|
|
"num_tokens": 9763472.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.400029230117798,
|
|
"epoch": 4.287142208087233,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003631241180480133,
|
|
"loss": 4.8249,
|
|
"mean_token_accuracy": 0.21257014870643615,
|
|
"num_tokens": 9773700.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.432605457305908,
|
|
"epoch": 4.291685597455702,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003627988161531845,
|
|
"loss": 4.8516,
|
|
"mean_token_accuracy": 0.2131851077079773,
|
|
"num_tokens": 9784042.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.4545575141906735,
|
|
"epoch": 4.296228986824171,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003624732976235046,
|
|
"loss": 4.9859,
|
|
"mean_token_accuracy": 0.21102089285850525,
|
|
"num_tokens": 9794464.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.417864799499512,
|
|
"epoch": 4.30077237619264,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003621475632621584,
|
|
"loss": 4.9131,
|
|
"mean_token_accuracy": 0.2079685002565384,
|
|
"num_tokens": 9804459.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.4460484981536865,
|
|
"epoch": 4.305315765561109,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003618216138728627,
|
|
"loss": 4.8911,
|
|
"mean_token_accuracy": 0.21270597726106644,
|
|
"num_tokens": 9814962.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.466338300704956,
|
|
"epoch": 4.309859154929577,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00036149545025986567,
|
|
"loss": 4.9451,
|
|
"mean_token_accuracy": 0.20737929493188859,
|
|
"num_tokens": 9825184.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.442379903793335,
|
|
"epoch": 4.314402544298046,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00036116907322794335,
|
|
"loss": 4.9845,
|
|
"mean_token_accuracy": 0.2021922692656517,
|
|
"num_tokens": 9835184.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.447183895111084,
|
|
"epoch": 4.318945933666515,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00036084248358239857,
|
|
"loss": 4.886,
|
|
"mean_token_accuracy": 0.20613111406564713,
|
|
"num_tokens": 9845358.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.4132002830505375,
|
|
"epoch": 4.323489323034984,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00036051568212905897,
|
|
"loss": 4.8462,
|
|
"mean_token_accuracy": 0.21071085929870606,
|
|
"num_tokens": 9856519.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.460366344451904,
|
|
"epoch": 4.328032712403453,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003601886696742745,
|
|
"loss": 4.9086,
|
|
"mean_token_accuracy": 0.20893526524305345,
|
|
"num_tokens": 9867478.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.39950008392334,
|
|
"epoch": 4.332576101771922,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000359861447024916,
|
|
"loss": 4.8895,
|
|
"mean_token_accuracy": 0.2118741661310196,
|
|
"num_tokens": 9877709.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.3682044506072994,
|
|
"epoch": 4.337119491140391,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00035953401498837285,
|
|
"loss": 4.8421,
|
|
"mean_token_accuracy": 0.21548546552658082,
|
|
"num_tokens": 9888134.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.4266761302947994,
|
|
"epoch": 4.34166288050886,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00035920637437255096,
|
|
"loss": 4.8874,
|
|
"mean_token_accuracy": 0.2128556713461876,
|
|
"num_tokens": 9899150.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.470839548110962,
|
|
"epoch": 4.346206269877328,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000358878525985871,
|
|
"loss": 4.8995,
|
|
"mean_token_accuracy": 0.20815165638923644,
|
|
"num_tokens": 9909599.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.420970821380616,
|
|
"epoch": 4.350749659245797,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003585504706372663,
|
|
"loss": 4.8568,
|
|
"mean_token_accuracy": 0.2151120513677597,
|
|
"num_tokens": 9919607.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.446282005310058,
|
|
"epoch": 4.355293048614266,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003582222091361808,
|
|
"loss": 4.9114,
|
|
"mean_token_accuracy": 0.2084006667137146,
|
|
"num_tokens": 9930431.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.3829092502594,
|
|
"epoch": 4.359836437982735,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003578937422925671,
|
|
"loss": 4.8815,
|
|
"mean_token_accuracy": 0.21002978831529617,
|
|
"num_tokens": 9940647.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.376507234573364,
|
|
"epoch": 4.364379827351204,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003575650709168846,
|
|
"loss": 4.9158,
|
|
"mean_token_accuracy": 0.2048715502023697,
|
|
"num_tokens": 9951110.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.438254499435425,
|
|
"epoch": 4.368923216719673,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00035723619582009715,
|
|
"loss": 4.8853,
|
|
"mean_token_accuracy": 0.20759367197752,
|
|
"num_tokens": 9961158.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.418239498138428,
|
|
"epoch": 4.373466606088142,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00035690711781367137,
|
|
"loss": 4.908,
|
|
"mean_token_accuracy": 0.20809367299079895,
|
|
"num_tokens": 9971051.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.345593881607056,
|
|
"epoch": 4.378009995456611,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003565778377095744,
|
|
"loss": 4.8897,
|
|
"mean_token_accuracy": 0.2013788029551506,
|
|
"num_tokens": 9981990.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.413355875015259,
|
|
"epoch": 4.38255338482508,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003562483563202724,
|
|
"loss": 4.8215,
|
|
"mean_token_accuracy": 0.2170127034187317,
|
|
"num_tokens": 9991668.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.376517295837402,
|
|
"epoch": 4.387096774193548,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00035591867445872784,
|
|
"loss": 4.7789,
|
|
"mean_token_accuracy": 0.21599389612674713,
|
|
"num_tokens": 10001519.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.357483291625977,
|
|
"epoch": 4.391640163562017,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000355588792938398,
|
|
"loss": 4.8564,
|
|
"mean_token_accuracy": 0.21310141086578369,
|
|
"num_tokens": 10010856.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.339485836029053,
|
|
"epoch": 4.396183552930486,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003552587125732326,
|
|
"loss": 4.8628,
|
|
"mean_token_accuracy": 0.21289247572422026,
|
|
"num_tokens": 10020647.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.399423599243164,
|
|
"epoch": 4.400726942298955,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00035492843417767224,
|
|
"loss": 4.8032,
|
|
"mean_token_accuracy": 0.21378337293863298,
|
|
"num_tokens": 10030114.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.445582008361816,
|
|
"epoch": 4.405270331667424,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000354597958566646,
|
|
"loss": 4.8733,
|
|
"mean_token_accuracy": 0.21809503585100173,
|
|
"num_tokens": 10039863.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.341729640960693,
|
|
"epoch": 4.409813721035893,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00035426728655556954,
|
|
"loss": 4.8288,
|
|
"mean_token_accuracy": 0.21136534959077835,
|
|
"num_tokens": 10049111.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.4089538097381595,
|
|
"epoch": 4.414357110404362,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003539364189603433,
|
|
"loss": 4.9205,
|
|
"mean_token_accuracy": 0.21054663211107255,
|
|
"num_tokens": 10058762.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.444057607650757,
|
|
"epoch": 4.418900499772831,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00035360535659735006,
|
|
"loss": 4.9123,
|
|
"mean_token_accuracy": 0.2066686823964119,
|
|
"num_tokens": 10068587.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.467214870452881,
|
|
"epoch": 4.423443889141299,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003532741002834534,
|
|
"loss": 4.8851,
|
|
"mean_token_accuracy": 0.20852230489253998,
|
|
"num_tokens": 10079386.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.389379453659058,
|
|
"epoch": 4.427987278509768,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00035294265083599526,
|
|
"loss": 4.8219,
|
|
"mean_token_accuracy": 0.2184127762913704,
|
|
"num_tokens": 10089490.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.36592059135437,
|
|
"epoch": 4.432530667878237,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003526110090727942,
|
|
"loss": 4.909,
|
|
"mean_token_accuracy": 0.20816168934106827,
|
|
"num_tokens": 10100465.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.436532592773437,
|
|
"epoch": 4.437074057246706,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003522791758121434,
|
|
"loss": 4.8333,
|
|
"mean_token_accuracy": 0.2150203615427017,
|
|
"num_tokens": 10110919.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.401555299758911,
|
|
"epoch": 4.441617446615175,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00035194715187280835,
|
|
"loss": 4.8786,
|
|
"mean_token_accuracy": 0.2135162338614464,
|
|
"num_tokens": 10121116.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.472104930877686,
|
|
"epoch": 4.446160835983644,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003516149380740252,
|
|
"loss": 4.9812,
|
|
"mean_token_accuracy": 0.209867961704731,
|
|
"num_tokens": 10132269.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.3837567329406735,
|
|
"epoch": 4.450704225352113,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003512825352354983,
|
|
"loss": 4.872,
|
|
"mean_token_accuracy": 0.21761986315250398,
|
|
"num_tokens": 10142856.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.348486757278442,
|
|
"epoch": 4.455247614720582,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00035094994417739884,
|
|
"loss": 4.8526,
|
|
"mean_token_accuracy": 0.21175325959920882,
|
|
"num_tokens": 10152377.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.466989040374756,
|
|
"epoch": 4.45979100408905,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000350617165720362,
|
|
"loss": 4.9266,
|
|
"mean_token_accuracy": 0.21066484302282334,
|
|
"num_tokens": 10161961.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.475636625289917,
|
|
"epoch": 4.464334393457519,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00035028420068548567,
|
|
"loss": 4.8836,
|
|
"mean_token_accuracy": 0.21329872757196428,
|
|
"num_tokens": 10171624.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.506438589096069,
|
|
"epoch": 4.468877782825988,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003499510498943279,
|
|
"loss": 4.9786,
|
|
"mean_token_accuracy": 0.20474951416254045,
|
|
"num_tokens": 10182642.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.396260786056518,
|
|
"epoch": 4.473421172194457,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00034961771416890526,
|
|
"loss": 4.9469,
|
|
"mean_token_accuracy": 0.2102530598640442,
|
|
"num_tokens": 10193307.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.40885443687439,
|
|
"epoch": 4.477964561562926,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003492841943316903,
|
|
"loss": 4.9241,
|
|
"mean_token_accuracy": 0.20724759250879288,
|
|
"num_tokens": 10203506.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.406883192062378,
|
|
"epoch": 4.482507950931395,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003489504912056103,
|
|
"loss": 4.8228,
|
|
"mean_token_accuracy": 0.2101452276110649,
|
|
"num_tokens": 10214266.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.297346353530884,
|
|
"epoch": 4.487051340299864,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003486166056140445,
|
|
"loss": 4.8126,
|
|
"mean_token_accuracy": 0.2176786720752716,
|
|
"num_tokens": 10224953.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.346699810028076,
|
|
"epoch": 4.491594729668333,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00034828253838082223,
|
|
"loss": 4.8594,
|
|
"mean_token_accuracy": 0.2050364449620247,
|
|
"num_tokens": 10235435.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.39209942817688,
|
|
"epoch": 4.496138119036801,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00034794829033022134,
|
|
"loss": 4.8661,
|
|
"mean_token_accuracy": 0.2144402965903282,
|
|
"num_tokens": 10245677.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.282089567184448,
|
|
"epoch": 4.50068150840527,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003476138622869657,
|
|
"loss": 4.8395,
|
|
"mean_token_accuracy": 0.20863480865955353,
|
|
"num_tokens": 10256135.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.3925806999206545,
|
|
"epoch": 4.505224897773739,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003472792550762232,
|
|
"loss": 4.9258,
|
|
"mean_token_accuracy": 0.20601883977651597,
|
|
"num_tokens": 10266465.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.507310104370117,
|
|
"epoch": 4.509768287142208,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003469444695236038,
|
|
"loss": 4.9323,
|
|
"mean_token_accuracy": 0.21040310114622116,
|
|
"num_tokens": 10278123.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.386733818054199,
|
|
"epoch": 4.514311676510677,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034660950645515765,
|
|
"loss": 4.894,
|
|
"mean_token_accuracy": 0.20797768533229827,
|
|
"num_tokens": 10287486.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.405367040634156,
|
|
"epoch": 4.518855065879146,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003462743666973728,
|
|
"loss": 4.8718,
|
|
"mean_token_accuracy": 0.2150442510843277,
|
|
"num_tokens": 10298100.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.519808387756347,
|
|
"epoch": 4.523398455247615,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003459390510771734,
|
|
"loss": 4.943,
|
|
"mean_token_accuracy": 0.20839779675006867,
|
|
"num_tokens": 10308549.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.400583553314209,
|
|
"epoch": 4.527941844616084,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003456035604219172,
|
|
"loss": 4.9007,
|
|
"mean_token_accuracy": 0.20511225908994674,
|
|
"num_tokens": 10319030.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.349795007705689,
|
|
"epoch": 4.532485233984552,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003452678955593942,
|
|
"loss": 4.9455,
|
|
"mean_token_accuracy": 0.2116200715303421,
|
|
"num_tokens": 10329613.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.454913330078125,
|
|
"epoch": 4.537028623353021,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000344932057317824,
|
|
"loss": 4.9546,
|
|
"mean_token_accuracy": 0.20076169520616532,
|
|
"num_tokens": 10339844.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.471723031997681,
|
|
"epoch": 4.54157201272149,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003445960465258542,
|
|
"loss": 4.9625,
|
|
"mean_token_accuracy": 0.2027738004922867,
|
|
"num_tokens": 10350906.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.54157201272149,
|
|
"eval_entropy": 5.247440436828968,
|
|
"eval_loss": 5.68981409072876,
|
|
"eval_mean_token_accuracy": 0.17128033904835235,
|
|
"eval_num_tokens": 10350906.0,
|
|
"eval_runtime": 2.0042,
|
|
"eval_samples_per_second": 1713.895,
|
|
"eval_steps_per_second": 214.549,
|
|
"step": 5000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 2230269845299200.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|