22277 lines
609 KiB
JSON
22277 lines
609 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 9.99136756019991,
|
|
"eval_steps": 500,
|
|
"global_step": 11000,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 5.337526845932007,
|
|
"epoch": 4.546115402089959,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003442598640125578,
|
|
"loss": 4.9073,
|
|
"mean_token_accuracy": 0.20096581876277925,
|
|
"num_tokens": 10361574.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.461961650848389,
|
|
"epoch": 4.550658791458428,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000343923510607432,
|
|
"loss": 4.9297,
|
|
"mean_token_accuracy": 0.20513624846935272,
|
|
"num_tokens": 10372714.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.338648843765259,
|
|
"epoch": 4.555202180826897,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003435869871403952,
|
|
"loss": 4.8306,
|
|
"mean_token_accuracy": 0.21388701498508453,
|
|
"num_tokens": 10382143.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.401988458633423,
|
|
"epoch": 4.559745570195366,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00034325029444178565,
|
|
"loss": 4.8696,
|
|
"mean_token_accuracy": 0.20959245264530182,
|
|
"num_tokens": 10392040.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.411912631988526,
|
|
"epoch": 4.564288959563835,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00034291343334235915,
|
|
"loss": 4.9438,
|
|
"mean_token_accuracy": 0.19913465082645415,
|
|
"num_tokens": 10403230.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.292964315414428,
|
|
"epoch": 4.568832348932304,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003425764046732869,
|
|
"loss": 4.8458,
|
|
"mean_token_accuracy": 0.2102683812379837,
|
|
"num_tokens": 10413466.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.384276437759399,
|
|
"epoch": 4.573375738300772,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00034223920926615376,
|
|
"loss": 4.8667,
|
|
"mean_token_accuracy": 0.21614075303077698,
|
|
"num_tokens": 10422559.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.418693017959595,
|
|
"epoch": 4.577919127669241,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00034190184795295577,
|
|
"loss": 4.8411,
|
|
"mean_token_accuracy": 0.2169215202331543,
|
|
"num_tokens": 10432544.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.4129682064056395,
|
|
"epoch": 4.58246251703771,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00034156432156609844,
|
|
"loss": 4.9292,
|
|
"mean_token_accuracy": 0.2027833104133606,
|
|
"num_tokens": 10442014.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.335578441619873,
|
|
"epoch": 4.587005906406179,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00034122663093839466,
|
|
"loss": 4.7728,
|
|
"mean_token_accuracy": 0.21595122814178466,
|
|
"num_tokens": 10451093.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.3574058532714846,
|
|
"epoch": 4.591549295774648,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003408887769030624,
|
|
"loss": 4.8058,
|
|
"mean_token_accuracy": 0.2119956523180008,
|
|
"num_tokens": 10461678.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.306751871109009,
|
|
"epoch": 4.596092685143117,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000340550760293723,
|
|
"loss": 4.8559,
|
|
"mean_token_accuracy": 0.21100128144025804,
|
|
"num_tokens": 10471607.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.4603527069091795,
|
|
"epoch": 4.600636074511586,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00034021258194439876,
|
|
"loss": 4.9208,
|
|
"mean_token_accuracy": 0.20606716871261596,
|
|
"num_tokens": 10482453.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.325381135940551,
|
|
"epoch": 4.605179463880054,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00033987424268951117,
|
|
"loss": 4.8062,
|
|
"mean_token_accuracy": 0.21982502043247223,
|
|
"num_tokens": 10492424.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.385643005371094,
|
|
"epoch": 4.609722853248523,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00033953574336387867,
|
|
"loss": 4.8391,
|
|
"mean_token_accuracy": 0.20995595902204514,
|
|
"num_tokens": 10500958.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.39823088645935,
|
|
"epoch": 4.614266242616992,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003391970848027148,
|
|
"loss": 4.9179,
|
|
"mean_token_accuracy": 0.2133391782641411,
|
|
"num_tokens": 10510906.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.387217330932617,
|
|
"epoch": 4.618809631985461,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003388582678416257,
|
|
"loss": 4.84,
|
|
"mean_token_accuracy": 0.21185368150472642,
|
|
"num_tokens": 10521832.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.374665641784668,
|
|
"epoch": 4.62335302135393,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00033851929331660876,
|
|
"loss": 4.8176,
|
|
"mean_token_accuracy": 0.21555525809526443,
|
|
"num_tokens": 10532310.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.342080783843994,
|
|
"epoch": 4.627896410722399,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00033818016206404983,
|
|
"loss": 4.8611,
|
|
"mean_token_accuracy": 0.20850327014923095,
|
|
"num_tokens": 10543979.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.5055166721344,
|
|
"epoch": 4.632439800090868,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003378408749207215,
|
|
"loss": 4.9608,
|
|
"mean_token_accuracy": 0.20202005952596663,
|
|
"num_tokens": 10553757.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.418175172805786,
|
|
"epoch": 4.636983189459337,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003375014327237812,
|
|
"loss": 4.9073,
|
|
"mean_token_accuracy": 0.20326604694128036,
|
|
"num_tokens": 10563575.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.441399383544922,
|
|
"epoch": 4.641526578827806,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003371618363107686,
|
|
"loss": 4.9348,
|
|
"mean_token_accuracy": 0.20748887956142426,
|
|
"num_tokens": 10572828.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.338515424728394,
|
|
"epoch": 4.646069968196274,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00033682208651960444,
|
|
"loss": 4.8768,
|
|
"mean_token_accuracy": 0.20922972708940507,
|
|
"num_tokens": 10582567.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.369612550735473,
|
|
"epoch": 4.650613357564743,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00033648218418858716,
|
|
"loss": 4.8816,
|
|
"mean_token_accuracy": 0.21072237193584442,
|
|
"num_tokens": 10594046.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.427084016799927,
|
|
"epoch": 4.655156746933212,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003361421301563922,
|
|
"loss": 4.8635,
|
|
"mean_token_accuracy": 0.21237473487854003,
|
|
"num_tokens": 10604343.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.3386157035827635,
|
|
"epoch": 4.659700136301681,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003358019252620691,
|
|
"loss": 4.9492,
|
|
"mean_token_accuracy": 0.20014230459928511,
|
|
"num_tokens": 10615140.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.431574916839599,
|
|
"epoch": 4.66424352567015,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00033546157034503954,
|
|
"loss": 4.8922,
|
|
"mean_token_accuracy": 0.210378298163414,
|
|
"num_tokens": 10624768.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.4156028747558596,
|
|
"epoch": 4.668786915038619,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003351210662450955,
|
|
"loss": 4.8068,
|
|
"mean_token_accuracy": 0.21552393287420274,
|
|
"num_tokens": 10635246.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.339911317825317,
|
|
"epoch": 4.673330304407088,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003347804138023968,
|
|
"loss": 4.8065,
|
|
"mean_token_accuracy": 0.22135637551546097,
|
|
"num_tokens": 10645118.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.357584238052368,
|
|
"epoch": 4.677873693775557,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003344396138574698,
|
|
"loss": 4.8379,
|
|
"mean_token_accuracy": 0.21505610942840575,
|
|
"num_tokens": 10655846.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.414635419845581,
|
|
"epoch": 4.682417083144026,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003340986672512042,
|
|
"loss": 4.9298,
|
|
"mean_token_accuracy": 0.20568588525056838,
|
|
"num_tokens": 10665741.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.428465986251831,
|
|
"epoch": 4.686960472512494,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003337575748248521,
|
|
"loss": 4.8948,
|
|
"mean_token_accuracy": 0.21208411157131196,
|
|
"num_tokens": 10676120.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.3839047908782955,
|
|
"epoch": 4.691503861880963,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033341633742002503,
|
|
"loss": 4.8859,
|
|
"mean_token_accuracy": 0.203281170129776,
|
|
"num_tokens": 10686195.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.361165380477905,
|
|
"epoch": 4.696047251249432,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003330749558786924,
|
|
"loss": 4.9148,
|
|
"mean_token_accuracy": 0.21346971690654754,
|
|
"num_tokens": 10697278.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.433807134628296,
|
|
"epoch": 4.700590640617901,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003327334310431791,
|
|
"loss": 4.9399,
|
|
"mean_token_accuracy": 0.21185509413480758,
|
|
"num_tokens": 10707937.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.405030298233032,
|
|
"epoch": 4.70513402998637,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003323917637561638,
|
|
"loss": 4.8013,
|
|
"mean_token_accuracy": 0.20688817501068116,
|
|
"num_tokens": 10717963.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.377948331832886,
|
|
"epoch": 4.709677419354839,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00033204995486067664,
|
|
"loss": 4.9012,
|
|
"mean_token_accuracy": 0.2065599277615547,
|
|
"num_tokens": 10728294.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.507592678070068,
|
|
"epoch": 4.714220808723308,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00033170800520009703,
|
|
"loss": 4.9411,
|
|
"mean_token_accuracy": 0.20752380639314652,
|
|
"num_tokens": 10739617.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.4214869976043705,
|
|
"epoch": 4.718764198091776,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003313659156181518,
|
|
"loss": 4.9552,
|
|
"mean_token_accuracy": 0.20658174455165862,
|
|
"num_tokens": 10750704.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.36991810798645,
|
|
"epoch": 4.723307587460245,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033102368695891283,
|
|
"loss": 4.846,
|
|
"mean_token_accuracy": 0.22009120881557465,
|
|
"num_tokens": 10760106.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.466967153549194,
|
|
"epoch": 4.727850976828714,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00033068132006679554,
|
|
"loss": 5.0063,
|
|
"mean_token_accuracy": 0.2034678190946579,
|
|
"num_tokens": 10770963.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.437391757965088,
|
|
"epoch": 4.732394366197183,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033033881578655605,
|
|
"loss": 4.9559,
|
|
"mean_token_accuracy": 0.20611633062362672,
|
|
"num_tokens": 10781441.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.459349203109741,
|
|
"epoch": 4.736937755565652,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003299961749632898,
|
|
"loss": 4.9281,
|
|
"mean_token_accuracy": 0.2148111268877983,
|
|
"num_tokens": 10791548.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.421885299682617,
|
|
"epoch": 4.741481144934121,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00032965339844242864,
|
|
"loss": 4.8806,
|
|
"mean_token_accuracy": 0.20914127081632614,
|
|
"num_tokens": 10801525.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.370760869979859,
|
|
"epoch": 4.74602453430259,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003293104870697399,
|
|
"loss": 4.8452,
|
|
"mean_token_accuracy": 0.2069530189037323,
|
|
"num_tokens": 10811907.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.427062702178955,
|
|
"epoch": 4.750567923671059,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00032896744169132305,
|
|
"loss": 4.8586,
|
|
"mean_token_accuracy": 0.217789526283741,
|
|
"num_tokens": 10822996.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.347399473190308,
|
|
"epoch": 4.755111313039528,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003286242631536087,
|
|
"loss": 4.8187,
|
|
"mean_token_accuracy": 0.2108131542801857,
|
|
"num_tokens": 10832561.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.4227358341217045,
|
|
"epoch": 4.759654702407996,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00032828095230335555,
|
|
"loss": 4.9147,
|
|
"mean_token_accuracy": 0.20365112870931626,
|
|
"num_tokens": 10842591.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.390651750564575,
|
|
"epoch": 4.764198091776465,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003279375099876491,
|
|
"loss": 4.9188,
|
|
"mean_token_accuracy": 0.20355417132377623,
|
|
"num_tokens": 10852888.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.475487184524536,
|
|
"epoch": 4.768741481144934,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003275939370538992,
|
|
"loss": 4.9393,
|
|
"mean_token_accuracy": 0.20861732512712478,
|
|
"num_tokens": 10863792.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.362081623077392,
|
|
"epoch": 4.773284870513403,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00032725023434983767,
|
|
"loss": 4.8014,
|
|
"mean_token_accuracy": 0.22023200243711472,
|
|
"num_tokens": 10873206.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.363426208496094,
|
|
"epoch": 4.777828259881872,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00032690640272351715,
|
|
"loss": 4.943,
|
|
"mean_token_accuracy": 0.20651764422655106,
|
|
"num_tokens": 10883686.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.312605667114258,
|
|
"epoch": 4.782371649250341,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003265624430233076,
|
|
"loss": 4.8711,
|
|
"mean_token_accuracy": 0.2175731763243675,
|
|
"num_tokens": 10894416.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.497582530975341,
|
|
"epoch": 4.78691503861881,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003262183560978956,
|
|
"loss": 4.9818,
|
|
"mean_token_accuracy": 0.21232758164405824,
|
|
"num_tokens": 10904405.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.507334804534912,
|
|
"epoch": 4.791458427987278,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003258741427962814,
|
|
"loss": 4.8942,
|
|
"mean_token_accuracy": 0.2103779450058937,
|
|
"num_tokens": 10915339.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.370587348937988,
|
|
"epoch": 4.796001817355747,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00032552980396777714,
|
|
"loss": 4.8727,
|
|
"mean_token_accuracy": 0.20776559859514238,
|
|
"num_tokens": 10926373.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.392536449432373,
|
|
"epoch": 4.800545206724216,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032518534046200464,
|
|
"loss": 4.9524,
|
|
"mean_token_accuracy": 0.2064978927373886,
|
|
"num_tokens": 10936695.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.490474891662598,
|
|
"epoch": 4.805088596092685,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003248407531288932,
|
|
"loss": 5.0153,
|
|
"mean_token_accuracy": 0.20103897899389267,
|
|
"num_tokens": 10949741.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.467194938659668,
|
|
"epoch": 4.809631985461154,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000324496042818678,
|
|
"loss": 4.9258,
|
|
"mean_token_accuracy": 0.21176911890506744,
|
|
"num_tokens": 10960786.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.39152455329895,
|
|
"epoch": 4.814175374829623,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00032415121038189746,
|
|
"loss": 4.8615,
|
|
"mean_token_accuracy": 0.21568342298269272,
|
|
"num_tokens": 10970423.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.372746801376342,
|
|
"epoch": 4.818718764198092,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00032380625666939125,
|
|
"loss": 4.884,
|
|
"mean_token_accuracy": 0.2131973996758461,
|
|
"num_tokens": 10980196.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.393996667861939,
|
|
"epoch": 4.823262153566561,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00032346118253229834,
|
|
"loss": 4.8991,
|
|
"mean_token_accuracy": 0.21001307666301727,
|
|
"num_tokens": 10989857.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.4272698879241945,
|
|
"epoch": 4.82780554293503,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00032311598882205496,
|
|
"loss": 4.8911,
|
|
"mean_token_accuracy": 0.20377996265888215,
|
|
"num_tokens": 11000276.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.442476987838745,
|
|
"epoch": 4.832348932303498,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00032277067639039216,
|
|
"loss": 4.9648,
|
|
"mean_token_accuracy": 0.19837054312229158,
|
|
"num_tokens": 11012502.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.416410207748413,
|
|
"epoch": 4.836892321671967,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032242524608933423,
|
|
"loss": 4.8265,
|
|
"mean_token_accuracy": 0.20706831514835358,
|
|
"num_tokens": 11023042.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.372754287719727,
|
|
"epoch": 4.841435711040436,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00032207969877119587,
|
|
"loss": 4.8742,
|
|
"mean_token_accuracy": 0.21531799733638762,
|
|
"num_tokens": 11032259.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.456787776947022,
|
|
"epoch": 4.845979100408905,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00032173403528858093,
|
|
"loss": 4.9996,
|
|
"mean_token_accuracy": 0.20667358487844467,
|
|
"num_tokens": 11042324.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.363194942474365,
|
|
"epoch": 4.850522489777374,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003213882564943796,
|
|
"loss": 4.9084,
|
|
"mean_token_accuracy": 0.2111344188451767,
|
|
"num_tokens": 11053519.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.4323410987854,
|
|
"epoch": 4.855065879145843,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003210423632417666,
|
|
"loss": 4.8786,
|
|
"mean_token_accuracy": 0.2140818491578102,
|
|
"num_tokens": 11063413.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.317426872253418,
|
|
"epoch": 4.859609268514312,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003206963563841993,
|
|
"loss": 4.8505,
|
|
"mean_token_accuracy": 0.20886150300502776,
|
|
"num_tokens": 11073232.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.352122640609741,
|
|
"epoch": 4.86415265788278,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00032035023677541515,
|
|
"loss": 4.8255,
|
|
"mean_token_accuracy": 0.20549337118864058,
|
|
"num_tokens": 11083408.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.45659031867981,
|
|
"epoch": 4.868696047251249,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003200040052694299,
|
|
"loss": 4.935,
|
|
"mean_token_accuracy": 0.2028595343232155,
|
|
"num_tokens": 11094125.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.423050260543823,
|
|
"epoch": 4.873239436619718,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00031965766272053534,
|
|
"loss": 4.9089,
|
|
"mean_token_accuracy": 0.20529586374759673,
|
|
"num_tokens": 11106293.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.379094219207763,
|
|
"epoch": 4.877782825988187,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00031931120998329737,
|
|
"loss": 4.8133,
|
|
"mean_token_accuracy": 0.22096642553806306,
|
|
"num_tokens": 11116485.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.4305956840515135,
|
|
"epoch": 4.882326215356656,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003189646479125536,
|
|
"loss": 4.9634,
|
|
"mean_token_accuracy": 0.20648754239082337,
|
|
"num_tokens": 11127876.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.447718334197998,
|
|
"epoch": 4.886869604725125,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003186179773634115,
|
|
"loss": 5.0351,
|
|
"mean_token_accuracy": 0.2046584367752075,
|
|
"num_tokens": 11138871.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.423292636871338,
|
|
"epoch": 4.891412994093594,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00031827119919124614,
|
|
"loss": 4.9215,
|
|
"mean_token_accuracy": 0.21052540689706803,
|
|
"num_tokens": 11148872.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.3357587337493895,
|
|
"epoch": 4.895956383462063,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00031792431425169826,
|
|
"loss": 4.7846,
|
|
"mean_token_accuracy": 0.21947864443063736,
|
|
"num_tokens": 11159716.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.408006763458252,
|
|
"epoch": 4.900499772830532,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00031757732340067195,
|
|
"loss": 4.8684,
|
|
"mean_token_accuracy": 0.21359502226114274,
|
|
"num_tokens": 11169187.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.457890176773072,
|
|
"epoch": 4.905043162199,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00031723022749433267,
|
|
"loss": 4.8309,
|
|
"mean_token_accuracy": 0.21711599826812744,
|
|
"num_tokens": 11179889.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.4713907718658445,
|
|
"epoch": 4.909586551567469,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000316883027389105,
|
|
"loss": 5.0123,
|
|
"mean_token_accuracy": 0.20325341075658798,
|
|
"num_tokens": 11189439.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.3972029209136965,
|
|
"epoch": 4.914129940935938,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003165357239416706,
|
|
"loss": 4.9267,
|
|
"mean_token_accuracy": 0.20619359016418456,
|
|
"num_tokens": 11200834.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.367896509170532,
|
|
"epoch": 4.918673330304407,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003161883180089665,
|
|
"loss": 4.8463,
|
|
"mean_token_accuracy": 0.2160806104540825,
|
|
"num_tokens": 11210475.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.397484874725341,
|
|
"epoch": 4.923216719672876,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003158408104481821,
|
|
"loss": 4.9175,
|
|
"mean_token_accuracy": 0.20287849456071855,
|
|
"num_tokens": 11220744.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.473115491867065,
|
|
"epoch": 4.927760109041345,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00031549320211675776,
|
|
"loss": 4.9992,
|
|
"mean_token_accuracy": 0.19899204075336457,
|
|
"num_tokens": 11231108.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.422712707519532,
|
|
"epoch": 4.932303498409814,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00031514549387238246,
|
|
"loss": 4.9288,
|
|
"mean_token_accuracy": 0.2115108013153076,
|
|
"num_tokens": 11241160.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.382764196395874,
|
|
"epoch": 4.936846887778283,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003147976865729918,
|
|
"loss": 4.9561,
|
|
"mean_token_accuracy": 0.2016743466258049,
|
|
"num_tokens": 11251711.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.407544708251953,
|
|
"epoch": 4.941390277146752,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00031444978107676567,
|
|
"loss": 4.9047,
|
|
"mean_token_accuracy": 0.21556849479675294,
|
|
"num_tokens": 11263025.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.41824049949646,
|
|
"epoch": 4.94593366651522,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00031410177824212635,
|
|
"loss": 4.9013,
|
|
"mean_token_accuracy": 0.2064465567469597,
|
|
"num_tokens": 11273927.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.400896310806274,
|
|
"epoch": 4.950477055883689,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003137536789277362,
|
|
"loss": 4.8906,
|
|
"mean_token_accuracy": 0.21240808069705963,
|
|
"num_tokens": 11284108.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.422491407394409,
|
|
"epoch": 4.955020445252158,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003134054839924957,
|
|
"loss": 4.8127,
|
|
"mean_token_accuracy": 0.21565574556589126,
|
|
"num_tokens": 11294530.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.344288682937622,
|
|
"epoch": 4.959563834620627,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00031305719429554124,
|
|
"loss": 4.873,
|
|
"mean_token_accuracy": 0.2016839161515236,
|
|
"num_tokens": 11304803.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.386518144607544,
|
|
"epoch": 4.964107223989096,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003127088106962429,
|
|
"loss": 4.9433,
|
|
"mean_token_accuracy": 0.21208882033824922,
|
|
"num_tokens": 11314567.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.4139063358306885,
|
|
"epoch": 4.968650613357565,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00031236033405420284,
|
|
"loss": 4.9077,
|
|
"mean_token_accuracy": 0.20902879983186723,
|
|
"num_tokens": 11324452.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.40776948928833,
|
|
"epoch": 4.973194002726034,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00031201176522925236,
|
|
"loss": 4.8973,
|
|
"mean_token_accuracy": 0.21027258336544036,
|
|
"num_tokens": 11334930.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.336411619186402,
|
|
"epoch": 4.977737392094502,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003116631050814503,
|
|
"loss": 4.8261,
|
|
"mean_token_accuracy": 0.20889938473701478,
|
|
"num_tokens": 11345487.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.443588972091675,
|
|
"epoch": 4.982280781462971,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00031131435447108116,
|
|
"loss": 4.9104,
|
|
"mean_token_accuracy": 0.2089818760752678,
|
|
"num_tokens": 11355150.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.449512100219726,
|
|
"epoch": 4.98682417083144,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003109655142586521,
|
|
"loss": 4.8837,
|
|
"mean_token_accuracy": 0.21147422194480897,
|
|
"num_tokens": 11366642.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.384327840805054,
|
|
"epoch": 4.991367560199909,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00031061658530489195,
|
|
"loss": 4.9104,
|
|
"mean_token_accuracy": 0.20778468251228333,
|
|
"num_tokens": 11377229.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.3677726745605465,
|
|
"epoch": 4.995910949568378,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00031026756847074804,
|
|
"loss": 4.9384,
|
|
"mean_token_accuracy": 0.2081951379776001,
|
|
"num_tokens": 11387444.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 4.995910949568378,
|
|
"eval_entropy": 5.256438230913739,
|
|
"eval_loss": 5.648169040679932,
|
|
"eval_mean_token_accuracy": 0.17371119156826373,
|
|
"eval_num_tokens": 11387444.0,
|
|
"eval_runtime": 1.9997,
|
|
"eval_samples_per_second": 1717.745,
|
|
"eval_steps_per_second": 215.031,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.397638585832384,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 0.0003099184646173849,
|
|
"loss": 4.8904,
|
|
"mean_token_accuracy": 0.21288909018039703,
|
|
"num_tokens": 11396985.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.402444314956665,
|
|
"epoch": 5.004543389368469,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00030956927460618126,
|
|
"loss": 4.6999,
|
|
"mean_token_accuracy": 0.21974647492170335,
|
|
"num_tokens": 11408747.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.377308797836304,
|
|
"epoch": 5.009086778736938,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003092199992987291,
|
|
"loss": 4.6229,
|
|
"mean_token_accuracy": 0.23138588219881057,
|
|
"num_tokens": 11418695.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.3167942523956295,
|
|
"epoch": 5.013630168105407,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00030887063955683044,
|
|
"loss": 4.6923,
|
|
"mean_token_accuracy": 0.2250524044036865,
|
|
"num_tokens": 11428143.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.4122966766357425,
|
|
"epoch": 5.018173557473876,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00030852119624249553,
|
|
"loss": 4.7357,
|
|
"mean_token_accuracy": 0.22295145094394683,
|
|
"num_tokens": 11439610.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.443834400177002,
|
|
"epoch": 5.022716946842344,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003081716702179412,
|
|
"loss": 4.6951,
|
|
"mean_token_accuracy": 0.22554124146699905,
|
|
"num_tokens": 11450035.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.325449419021607,
|
|
"epoch": 5.027260336210813,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000307822062345588,
|
|
"loss": 4.6678,
|
|
"mean_token_accuracy": 0.234792061150074,
|
|
"num_tokens": 11460736.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.3676934242248535,
|
|
"epoch": 5.031803725579282,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003074723734880588,
|
|
"loss": 4.6546,
|
|
"mean_token_accuracy": 0.22833489775657653,
|
|
"num_tokens": 11471319.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.333033895492553,
|
|
"epoch": 5.036347114947751,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000307122604508176,
|
|
"loss": 4.6557,
|
|
"mean_token_accuracy": 0.22927162945270538,
|
|
"num_tokens": 11481052.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.307074546813965,
|
|
"epoch": 5.04089050431622,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003067727562689598,
|
|
"loss": 4.652,
|
|
"mean_token_accuracy": 0.22286555171012878,
|
|
"num_tokens": 11491166.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.383533000946045,
|
|
"epoch": 5.045433893684689,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003064228296336259,
|
|
"loss": 4.6402,
|
|
"mean_token_accuracy": 0.22903617322444916,
|
|
"num_tokens": 11501007.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.368576192855835,
|
|
"epoch": 5.049977283053158,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003060728254655836,
|
|
"loss": 4.7165,
|
|
"mean_token_accuracy": 0.22236028164625168,
|
|
"num_tokens": 11512197.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.440391397476196,
|
|
"epoch": 5.054520672421627,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00030572274462843327,
|
|
"loss": 4.7021,
|
|
"mean_token_accuracy": 0.21848050653934478,
|
|
"num_tokens": 11522396.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.384906482696533,
|
|
"epoch": 5.059064061790095,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00030537258798596465,
|
|
"loss": 4.7173,
|
|
"mean_token_accuracy": 0.22784159630537032,
|
|
"num_tokens": 11533034.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.456916332244873,
|
|
"epoch": 5.063607451158564,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003050223564021543,
|
|
"loss": 4.7224,
|
|
"mean_token_accuracy": 0.22801146060228347,
|
|
"num_tokens": 11543281.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.348969459533691,
|
|
"epoch": 5.068150840527033,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003046720507411641,
|
|
"loss": 4.6722,
|
|
"mean_token_accuracy": 0.22941800504922866,
|
|
"num_tokens": 11552905.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.325329399108886,
|
|
"epoch": 5.072694229895502,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003043216718673383,
|
|
"loss": 4.6731,
|
|
"mean_token_accuracy": 0.23223626166582106,
|
|
"num_tokens": 11562397.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.4154332160949705,
|
|
"epoch": 5.077237619263971,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00030397122064520186,
|
|
"loss": 4.7492,
|
|
"mean_token_accuracy": 0.21953937411308289,
|
|
"num_tokens": 11573375.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.4002808094024655,
|
|
"epoch": 5.08178100863244,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00030362069793945856,
|
|
"loss": 4.6834,
|
|
"mean_token_accuracy": 0.2266400396823883,
|
|
"num_tokens": 11584155.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.341732406616211,
|
|
"epoch": 5.086324398000909,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003032701046149881,
|
|
"loss": 4.6725,
|
|
"mean_token_accuracy": 0.22485776394605636,
|
|
"num_tokens": 11593929.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.394675636291504,
|
|
"epoch": 5.090867787369378,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003029194415368448,
|
|
"loss": 4.7012,
|
|
"mean_token_accuracy": 0.23234383314847945,
|
|
"num_tokens": 11604891.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.39636492729187,
|
|
"epoch": 5.095411176737846,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003025687095702548,
|
|
"loss": 4.7441,
|
|
"mean_token_accuracy": 0.2124445840716362,
|
|
"num_tokens": 11616377.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.29693570137024,
|
|
"epoch": 5.099954566106315,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003022179095806146,
|
|
"loss": 4.6998,
|
|
"mean_token_accuracy": 0.22328357696533202,
|
|
"num_tokens": 11627079.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.356540060043335,
|
|
"epoch": 5.104497955474784,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000301867042433488,
|
|
"loss": 4.6937,
|
|
"mean_token_accuracy": 0.2276850998401642,
|
|
"num_tokens": 11638014.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.45648226737976,
|
|
"epoch": 5.109041344843253,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00030151610899460505,
|
|
"loss": 4.7974,
|
|
"mean_token_accuracy": 0.228281170129776,
|
|
"num_tokens": 11648759.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.4036558151245115,
|
|
"epoch": 5.113584734211722,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00030116511012985916,
|
|
"loss": 4.7104,
|
|
"mean_token_accuracy": 0.22886117547750473,
|
|
"num_tokens": 11659050.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.386154842376709,
|
|
"epoch": 5.118128123580191,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003008140467053048,
|
|
"loss": 4.7359,
|
|
"mean_token_accuracy": 0.22094981968402863,
|
|
"num_tokens": 11670131.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.281435871124268,
|
|
"epoch": 5.12267151294866,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003004629195871565,
|
|
"loss": 4.6363,
|
|
"mean_token_accuracy": 0.22783708423376084,
|
|
"num_tokens": 11680564.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.4245960235595705,
|
|
"epoch": 5.127214902317129,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003001117296417852,
|
|
"loss": 4.7694,
|
|
"mean_token_accuracy": 0.21815529465675354,
|
|
"num_tokens": 11691910.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.332416963577271,
|
|
"epoch": 5.131758291685597,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002997604777357173,
|
|
"loss": 4.7015,
|
|
"mean_token_accuracy": 0.22673041820526124,
|
|
"num_tokens": 11702174.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.411578226089477,
|
|
"epoch": 5.136301681054066,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000299409164735632,
|
|
"loss": 4.8151,
|
|
"mean_token_accuracy": 0.22040149569511414,
|
|
"num_tokens": 11712841.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.408714389801025,
|
|
"epoch": 5.140845070422535,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002990577915083591,
|
|
"loss": 4.7168,
|
|
"mean_token_accuracy": 0.22699419558048248,
|
|
"num_tokens": 11722781.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.352818250656128,
|
|
"epoch": 5.145388459791004,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00029870635892087716,
|
|
"loss": 4.6152,
|
|
"mean_token_accuracy": 0.2327147752046585,
|
|
"num_tokens": 11732774.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.299087333679199,
|
|
"epoch": 5.149931849159473,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002983548678403112,
|
|
"loss": 4.7021,
|
|
"mean_token_accuracy": 0.22236208915710448,
|
|
"num_tokens": 11743308.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.38860936164856,
|
|
"epoch": 5.154475238527942,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002980033191339305,
|
|
"loss": 4.7787,
|
|
"mean_token_accuracy": 0.2148905485868454,
|
|
"num_tokens": 11754572.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.315798664093018,
|
|
"epoch": 5.159018627896411,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002976517136691465,
|
|
"loss": 4.6973,
|
|
"mean_token_accuracy": 0.22461562156677245,
|
|
"num_tokens": 11764406.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.344510126113891,
|
|
"epoch": 5.16356201726488,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00029730005231351074,
|
|
"loss": 4.6811,
|
|
"mean_token_accuracy": 0.22377657741308213,
|
|
"num_tokens": 11774121.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.382678604125976,
|
|
"epoch": 5.168105406633348,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002969483359347127,
|
|
"loss": 4.7315,
|
|
"mean_token_accuracy": 0.2224814310669899,
|
|
"num_tokens": 11784820.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.309485626220703,
|
|
"epoch": 5.172648796001817,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002965965654005776,
|
|
"loss": 4.6922,
|
|
"mean_token_accuracy": 0.22986993044614792,
|
|
"num_tokens": 11794891.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.359425401687622,
|
|
"epoch": 5.177192185370286,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002962447415790642,
|
|
"loss": 4.6647,
|
|
"mean_token_accuracy": 0.22956188321113585,
|
|
"num_tokens": 11804842.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.349454593658447,
|
|
"epoch": 5.181735574738755,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00029589286533826276,
|
|
"loss": 4.7228,
|
|
"mean_token_accuracy": 0.2239353120326996,
|
|
"num_tokens": 11815893.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.317618560791016,
|
|
"epoch": 5.186278964107224,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002955409375463931,
|
|
"loss": 4.6342,
|
|
"mean_token_accuracy": 0.23248186856508254,
|
|
"num_tokens": 11826332.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.380689287185669,
|
|
"epoch": 5.190822353475693,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002951889590718019,
|
|
"loss": 4.7599,
|
|
"mean_token_accuracy": 0.21811264604330063,
|
|
"num_tokens": 11836977.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.33549656867981,
|
|
"epoch": 5.195365742844162,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002948369307829612,
|
|
"loss": 4.6758,
|
|
"mean_token_accuracy": 0.2288920983672142,
|
|
"num_tokens": 11846483.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.378067874908448,
|
|
"epoch": 5.199909132212631,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002944848535484656,
|
|
"loss": 4.7301,
|
|
"mean_token_accuracy": 0.22489935904741287,
|
|
"num_tokens": 11857055.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.3050370693206785,
|
|
"epoch": 5.2044525215811,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00029413272823703085,
|
|
"loss": 4.6231,
|
|
"mean_token_accuracy": 0.23340309411287308,
|
|
"num_tokens": 11866929.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.332051992416382,
|
|
"epoch": 5.208995910949568,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002937805557174911,
|
|
"loss": 4.7318,
|
|
"mean_token_accuracy": 0.22128495424985886,
|
|
"num_tokens": 11875973.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.377699708938598,
|
|
"epoch": 5.213539300318037,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00029342833685879713,
|
|
"loss": 4.7383,
|
|
"mean_token_accuracy": 0.23146028220653533,
|
|
"num_tokens": 11886273.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.446506071090698,
|
|
"epoch": 5.218082689686506,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002930760725300137,
|
|
"loss": 4.7676,
|
|
"mean_token_accuracy": 0.22437227368354798,
|
|
"num_tokens": 11897462.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.376716566085816,
|
|
"epoch": 5.222626079054975,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00029272376360031833,
|
|
"loss": 4.7175,
|
|
"mean_token_accuracy": 0.22333021461963654,
|
|
"num_tokens": 11908116.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.390292739868164,
|
|
"epoch": 5.227169468423444,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00029237141093899804,
|
|
"loss": 4.7761,
|
|
"mean_token_accuracy": 0.2183089569211006,
|
|
"num_tokens": 11918990.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.39310736656189,
|
|
"epoch": 5.231712857791913,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00029201901541544824,
|
|
"loss": 4.7625,
|
|
"mean_token_accuracy": 0.22316605895757674,
|
|
"num_tokens": 11929201.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.323505592346192,
|
|
"epoch": 5.236256247160382,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002916665778991696,
|
|
"loss": 4.7285,
|
|
"mean_token_accuracy": 0.21832881718873978,
|
|
"num_tokens": 11939002.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.3333453178405765,
|
|
"epoch": 5.240799636528851,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002913140992597668,
|
|
"loss": 4.6114,
|
|
"mean_token_accuracy": 0.2301402509212494,
|
|
"num_tokens": 11949255.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.35742826461792,
|
|
"epoch": 5.245343025897319,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002909615803669458,
|
|
"loss": 4.6815,
|
|
"mean_token_accuracy": 0.2295151948928833,
|
|
"num_tokens": 11959108.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.237590360641479,
|
|
"epoch": 5.249886415265788,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002906090220905119,
|
|
"loss": 4.6099,
|
|
"mean_token_accuracy": 0.23081846088171004,
|
|
"num_tokens": 11969196.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.291966676712036,
|
|
"epoch": 5.254429804634257,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002902564253003677,
|
|
"loss": 4.647,
|
|
"mean_token_accuracy": 0.23048539012670516,
|
|
"num_tokens": 11979101.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.353731536865235,
|
|
"epoch": 5.258973194002726,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002899037908665106,
|
|
"loss": 4.6956,
|
|
"mean_token_accuracy": 0.22405403554439546,
|
|
"num_tokens": 11988396.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.355870056152344,
|
|
"epoch": 5.263516583371195,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002895511196590312,
|
|
"loss": 4.6872,
|
|
"mean_token_accuracy": 0.2304904043674469,
|
|
"num_tokens": 11998355.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.31823616027832,
|
|
"epoch": 5.268059972739664,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002891984125481104,
|
|
"loss": 4.6797,
|
|
"mean_token_accuracy": 0.23322721272706987,
|
|
"num_tokens": 12008938.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.373196744918824,
|
|
"epoch": 5.272603362108133,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00028884567040401814,
|
|
"loss": 4.7742,
|
|
"mean_token_accuracy": 0.21823119670152663,
|
|
"num_tokens": 12019875.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.378669691085816,
|
|
"epoch": 5.277146751476602,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002884928940971105,
|
|
"loss": 4.7126,
|
|
"mean_token_accuracy": 0.226912958920002,
|
|
"num_tokens": 12029368.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.351129198074341,
|
|
"epoch": 5.28169014084507,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000288140084497828,
|
|
"loss": 4.6712,
|
|
"mean_token_accuracy": 0.22691742032766343,
|
|
"num_tokens": 12040191.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.293002653121948,
|
|
"epoch": 5.286233530213539,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002877872424766932,
|
|
"loss": 4.6599,
|
|
"mean_token_accuracy": 0.2331748604774475,
|
|
"num_tokens": 12050087.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.2632646560668945,
|
|
"epoch": 5.290776919582008,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002874343689043086,
|
|
"loss": 4.6878,
|
|
"mean_token_accuracy": 0.23522936403751374,
|
|
"num_tokens": 12061014.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.3365601062774655,
|
|
"epoch": 5.295320308950477,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000287081464651355,
|
|
"loss": 4.7109,
|
|
"mean_token_accuracy": 0.23040277659893035,
|
|
"num_tokens": 12071055.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.33727240562439,
|
|
"epoch": 5.299863698318946,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002867285305885882,
|
|
"loss": 4.711,
|
|
"mean_token_accuracy": 0.23001304864883423,
|
|
"num_tokens": 12081323.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.295901250839234,
|
|
"epoch": 5.304407087687415,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00028637556758683816,
|
|
"loss": 4.6758,
|
|
"mean_token_accuracy": 0.23072714358568192,
|
|
"num_tokens": 12091482.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.2397970199584964,
|
|
"epoch": 5.308950477055884,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002860225765170058,
|
|
"loss": 4.5964,
|
|
"mean_token_accuracy": 0.23994662314653398,
|
|
"num_tokens": 12101960.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.321013498306274,
|
|
"epoch": 5.313493866424353,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00028566955825006174,
|
|
"loss": 4.7003,
|
|
"mean_token_accuracy": 0.22388923466205596,
|
|
"num_tokens": 12111654.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.43159670829773,
|
|
"epoch": 5.318037255792821,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00028531651365704325,
|
|
"loss": 4.7967,
|
|
"mean_token_accuracy": 0.22098034024238586,
|
|
"num_tokens": 12123096.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.334747409820556,
|
|
"epoch": 5.32258064516129,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00028496344360905287,
|
|
"loss": 4.6854,
|
|
"mean_token_accuracy": 0.225448676943779,
|
|
"num_tokens": 12133064.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.321415185928345,
|
|
"epoch": 5.327124034529759,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00028461034897725587,
|
|
"loss": 4.708,
|
|
"mean_token_accuracy": 0.22682421654462814,
|
|
"num_tokens": 12144386.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.343523168563843,
|
|
"epoch": 5.331667423898228,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00028425723063287803,
|
|
"loss": 4.7711,
|
|
"mean_token_accuracy": 0.22161225974559784,
|
|
"num_tokens": 12156111.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.425982284545898,
|
|
"epoch": 5.336210813266697,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00028390408944720385,
|
|
"loss": 4.7952,
|
|
"mean_token_accuracy": 0.21704222410917282,
|
|
"num_tokens": 12166462.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.3578962802886965,
|
|
"epoch": 5.340754202635166,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002835509262915741,
|
|
"loss": 4.8139,
|
|
"mean_token_accuracy": 0.21395985633134842,
|
|
"num_tokens": 12178635.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.298470211029053,
|
|
"epoch": 5.345297592003635,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00028319774203738375,
|
|
"loss": 4.6703,
|
|
"mean_token_accuracy": 0.22469234019517897,
|
|
"num_tokens": 12188880.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.304668712615967,
|
|
"epoch": 5.349840981372104,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002828445375560798,
|
|
"loss": 4.6807,
|
|
"mean_token_accuracy": 0.22120783776044844,
|
|
"num_tokens": 12199733.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.312647771835327,
|
|
"epoch": 5.354384370740572,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00028249131371915927,
|
|
"loss": 4.7313,
|
|
"mean_token_accuracy": 0.22074017822742462,
|
|
"num_tokens": 12209925.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.474558639526367,
|
|
"epoch": 5.358927760109041,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002821380713981668,
|
|
"loss": 4.7753,
|
|
"mean_token_accuracy": 0.22338093519210817,
|
|
"num_tokens": 12219870.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.330374717712402,
|
|
"epoch": 5.36347114947751,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00028178481146469273,
|
|
"loss": 4.6972,
|
|
"mean_token_accuracy": 0.23345100581645967,
|
|
"num_tokens": 12230295.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.34260048866272,
|
|
"epoch": 5.368014538845979,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002814315347903709,
|
|
"loss": 4.7165,
|
|
"mean_token_accuracy": 0.23311403393745422,
|
|
"num_tokens": 12239700.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.336610174179077,
|
|
"epoch": 5.372557928214448,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002810782422468763,
|
|
"loss": 4.7668,
|
|
"mean_token_accuracy": 0.22191945165395738,
|
|
"num_tokens": 12249529.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.332798290252685,
|
|
"epoch": 5.377101317582917,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00028072493470592317,
|
|
"loss": 4.6604,
|
|
"mean_token_accuracy": 0.22941354662179947,
|
|
"num_tokens": 12259774.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.368777084350586,
|
|
"epoch": 5.381644706951386,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00028037161303926265,
|
|
"loss": 4.7306,
|
|
"mean_token_accuracy": 0.22430871725082396,
|
|
"num_tokens": 12270484.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.33012900352478,
|
|
"epoch": 5.386188096319855,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000280018278118681,
|
|
"loss": 4.6839,
|
|
"mean_token_accuracy": 0.22534373998641968,
|
|
"num_tokens": 12281195.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.2529988288879395,
|
|
"epoch": 5.390731485688324,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000279664930815997,
|
|
"loss": 4.6868,
|
|
"mean_token_accuracy": 0.22569240927696227,
|
|
"num_tokens": 12292052.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.3173116683959964,
|
|
"epoch": 5.395274875056792,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002793115720030599,
|
|
"loss": 4.7639,
|
|
"mean_token_accuracy": 0.2249886929988861,
|
|
"num_tokens": 12302562.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.375454950332641,
|
|
"epoch": 5.399818264425261,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002789582025517474,
|
|
"loss": 4.7345,
|
|
"mean_token_accuracy": 0.22930043637752534,
|
|
"num_tokens": 12313875.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.380015420913696,
|
|
"epoch": 5.40436165379373,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002786048233339637,
|
|
"loss": 4.7411,
|
|
"mean_token_accuracy": 0.22127102613449096,
|
|
"num_tokens": 12324500.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.334359073638916,
|
|
"epoch": 5.408905043162199,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00027825143522163675,
|
|
"loss": 4.7208,
|
|
"mean_token_accuracy": 0.22637515217065812,
|
|
"num_tokens": 12335474.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.331499195098877,
|
|
"epoch": 5.413448432530668,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002778980390867165,
|
|
"loss": 4.6848,
|
|
"mean_token_accuracy": 0.2249290391802788,
|
|
"num_tokens": 12346174.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.370951223373413,
|
|
"epoch": 5.417991821899137,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00027754463580117306,
|
|
"loss": 4.6999,
|
|
"mean_token_accuracy": 0.2331499397754669,
|
|
"num_tokens": 12356108.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.354803228378296,
|
|
"epoch": 5.422535211267606,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00027719122623699353,
|
|
"loss": 4.7462,
|
|
"mean_token_accuracy": 0.22911420166492463,
|
|
"num_tokens": 12366699.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.435253763198853,
|
|
"epoch": 5.427078600636074,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002768378112661813,
|
|
"loss": 4.7567,
|
|
"mean_token_accuracy": 0.2168062224984169,
|
|
"num_tokens": 12377225.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 5.425332498550415,
|
|
"epoch": 5.431621990004543,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002764843917607523,
|
|
"loss": 4.7687,
|
|
"mean_token_accuracy": 0.2263517364859581,
|
|
"num_tokens": 12388936.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.308147048950195,
|
|
"epoch": 5.436165379373012,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002761309685927343,
|
|
"loss": 4.6959,
|
|
"mean_token_accuracy": 0.2217288002371788,
|
|
"num_tokens": 12399130.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.34355058670044,
|
|
"epoch": 5.440708768741481,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002757775426341636,
|
|
"loss": 4.6869,
|
|
"mean_token_accuracy": 0.23334074020385742,
|
|
"num_tokens": 12410181.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.349674987792969,
|
|
"epoch": 5.44525215810995,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00027542411475708385,
|
|
"loss": 4.7859,
|
|
"mean_token_accuracy": 0.21799193173646927,
|
|
"num_tokens": 12421173.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.408291625976562,
|
|
"epoch": 5.449795547478419,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00027507068583354304,
|
|
"loss": 4.7972,
|
|
"mean_token_accuracy": 0.2104117438197136,
|
|
"num_tokens": 12433195.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 5.449795547478419,
|
|
"eval_entropy": 5.245359480658243,
|
|
"eval_loss": 5.673709392547607,
|
|
"eval_mean_token_accuracy": 0.17348870330078658,
|
|
"eval_num_tokens": 12433195.0,
|
|
"eval_runtime": 2.0057,
|
|
"eval_samples_per_second": 1712.633,
|
|
"eval_steps_per_second": 214.391,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 5.358231735229492,
|
|
"epoch": 5.454338936846888,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00027471725673559205,
|
|
"loss": 4.706,
|
|
"mean_token_accuracy": 0.2219674453139305,
|
|
"num_tokens": 12442944.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 5.321769189834595,
|
|
"epoch": 5.458882326215357,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00027436382833528184,
|
|
"loss": 4.6723,
|
|
"mean_token_accuracy": 0.22999544441699982,
|
|
"num_tokens": 12453133.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 5.30899133682251,
|
|
"epoch": 5.463425715583826,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002740104015046621,
|
|
"loss": 4.7841,
|
|
"mean_token_accuracy": 0.21290091574192047,
|
|
"num_tokens": 12464606.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 5.310091733932495,
|
|
"epoch": 5.467969104952294,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00027365697711577815,
|
|
"loss": 4.6789,
|
|
"mean_token_accuracy": 0.22543101459741594,
|
|
"num_tokens": 12474490.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 5.378962182998658,
|
|
"epoch": 5.472512494320763,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002733035560406697,
|
|
"loss": 4.6897,
|
|
"mean_token_accuracy": 0.23378607332706453,
|
|
"num_tokens": 12485158.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 5.2819640159606935,
|
|
"epoch": 5.477055883689232,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00027295013915136805,
|
|
"loss": 4.7381,
|
|
"mean_token_accuracy": 0.22038409411907195,
|
|
"num_tokens": 12496265.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 5.287884950637817,
|
|
"epoch": 5.481599273057701,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00027259672731989436,
|
|
"loss": 4.6176,
|
|
"mean_token_accuracy": 0.23236948996782303,
|
|
"num_tokens": 12506719.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 5.356770467758179,
|
|
"epoch": 5.48614266242617,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000272243321418257,
|
|
"loss": 4.7022,
|
|
"mean_token_accuracy": 0.22641584873199463,
|
|
"num_tokens": 12515990.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 5.339677047729492,
|
|
"epoch": 5.490686051794639,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002718899223184501,
|
|
"loss": 4.7292,
|
|
"mean_token_accuracy": 0.22442852407693864,
|
|
"num_tokens": 12526557.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 5.289336633682251,
|
|
"epoch": 5.495229441163108,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00027153653089245067,
|
|
"loss": 4.6565,
|
|
"mean_token_accuracy": 0.23662521094083785,
|
|
"num_tokens": 12536657.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 5.289770698547363,
|
|
"epoch": 5.499772830531577,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000271183148012217,
|
|
"loss": 4.6742,
|
|
"mean_token_accuracy": 0.22677171379327773,
|
|
"num_tokens": 12546958.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 5.298456192016602,
|
|
"epoch": 5.504316219900045,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002708297745496861,
|
|
"loss": 4.7033,
|
|
"mean_token_accuracy": 0.22284281849861146,
|
|
"num_tokens": 12556978.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 5.319316864013672,
|
|
"epoch": 5.508859609268514,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00027047641137677213,
|
|
"loss": 4.6643,
|
|
"mean_token_accuracy": 0.2302199497818947,
|
|
"num_tokens": 12566954.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 5.339248132705689,
|
|
"epoch": 5.513402998636983,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002701230593653632,
|
|
"loss": 4.7143,
|
|
"mean_token_accuracy": 0.2293867975473404,
|
|
"num_tokens": 12577129.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 5.358866596221924,
|
|
"epoch": 5.517946388005452,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002697697193873207,
|
|
"loss": 4.7035,
|
|
"mean_token_accuracy": 0.21841893047094346,
|
|
"num_tokens": 12586802.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 5.361600112915039,
|
|
"epoch": 5.522489777373921,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00026941639231447547,
|
|
"loss": 4.7603,
|
|
"mean_token_accuracy": 0.221048504114151,
|
|
"num_tokens": 12597732.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 5.347931289672852,
|
|
"epoch": 5.52703316674239,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002690630790186273,
|
|
"loss": 4.6319,
|
|
"mean_token_accuracy": 0.2292472243309021,
|
|
"num_tokens": 12607185.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 5.3770088195800785,
|
|
"epoch": 5.531576556110859,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00026870978037154137,
|
|
"loss": 4.7401,
|
|
"mean_token_accuracy": 0.22142680883407592,
|
|
"num_tokens": 12616821.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 5.332234573364258,
|
|
"epoch": 5.536119945479328,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002683564972449472,
|
|
"loss": 4.7706,
|
|
"mean_token_accuracy": 0.22039098888635636,
|
|
"num_tokens": 12626698.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 5.324936723709106,
|
|
"epoch": 5.540663334847796,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002680032305105355,
|
|
"loss": 4.6877,
|
|
"mean_token_accuracy": 0.2317501425743103,
|
|
"num_tokens": 12636588.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 5.400929498672485,
|
|
"epoch": 5.545206724216265,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002676499810399572,
|
|
"loss": 4.7531,
|
|
"mean_token_accuracy": 0.22503160536289216,
|
|
"num_tokens": 12647196.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 5.340239858627319,
|
|
"epoch": 5.549750113584734,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00026729674970481985,
|
|
"loss": 4.7307,
|
|
"mean_token_accuracy": 0.22162305265665055,
|
|
"num_tokens": 12658631.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 5.324551010131836,
|
|
"epoch": 5.554293502953203,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00026694353737668705,
|
|
"loss": 4.7651,
|
|
"mean_token_accuracy": 0.22232725024223327,
|
|
"num_tokens": 12668905.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 5.3123087882995605,
|
|
"epoch": 5.558836892321672,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00026659034492707473,
|
|
"loss": 4.6915,
|
|
"mean_token_accuracy": 0.22733392417430878,
|
|
"num_tokens": 12678270.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 5.3269096374511715,
|
|
"epoch": 5.563380281690141,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002662371732274504,
|
|
"loss": 4.7064,
|
|
"mean_token_accuracy": 0.2272703930735588,
|
|
"num_tokens": 12688248.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 5.347763442993164,
|
|
"epoch": 5.56792367105861,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00026588402314923,
|
|
"loss": 4.7934,
|
|
"mean_token_accuracy": 0.22125670909881592,
|
|
"num_tokens": 12698418.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 5.2642481327056885,
|
|
"epoch": 5.572467060427079,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002655308955637763,
|
|
"loss": 4.6267,
|
|
"mean_token_accuracy": 0.22882166057825087,
|
|
"num_tokens": 12708578.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 5.319665908813477,
|
|
"epoch": 5.577010449795548,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002651777913423965,
|
|
"loss": 4.7462,
|
|
"mean_token_accuracy": 0.2205069974064827,
|
|
"num_tokens": 12718899.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 5.304476165771485,
|
|
"epoch": 5.581553839164016,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002648247113563402,
|
|
"loss": 4.6745,
|
|
"mean_token_accuracy": 0.22488473802804948,
|
|
"num_tokens": 12728342.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 5.430471515655517,
|
|
"epoch": 5.586097228532485,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00026447165647679703,
|
|
"loss": 4.7466,
|
|
"mean_token_accuracy": 0.23093189746141435,
|
|
"num_tokens": 12739337.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 5.319414234161377,
|
|
"epoch": 5.590640617900954,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000264118627574895,
|
|
"loss": 4.7215,
|
|
"mean_token_accuracy": 0.22540283650159837,
|
|
"num_tokens": 12749650.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 5.341633701324463,
|
|
"epoch": 5.595184007269423,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00026376562552169776,
|
|
"loss": 4.7778,
|
|
"mean_token_accuracy": 0.21634038239717485,
|
|
"num_tokens": 12760573.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 5.470093870162964,
|
|
"epoch": 5.599727396637892,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00026341265118820284,
|
|
"loss": 4.8008,
|
|
"mean_token_accuracy": 0.21790637522935868,
|
|
"num_tokens": 12771070.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 5.38938045501709,
|
|
"epoch": 5.604270786006361,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00026305970544533903,
|
|
"loss": 4.7022,
|
|
"mean_token_accuracy": 0.227985155582428,
|
|
"num_tokens": 12781896.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 5.319741487503052,
|
|
"epoch": 5.60881417537483,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00026270678916396527,
|
|
"loss": 4.7537,
|
|
"mean_token_accuracy": 0.22228948324918746,
|
|
"num_tokens": 12792675.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 5.315591049194336,
|
|
"epoch": 5.613357564743298,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00026235390321486705,
|
|
"loss": 4.6608,
|
|
"mean_token_accuracy": 0.23118897825479506,
|
|
"num_tokens": 12802371.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 5.404679250717163,
|
|
"epoch": 5.617900954111767,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00026200104846875574,
|
|
"loss": 4.7271,
|
|
"mean_token_accuracy": 0.22576195299625396,
|
|
"num_tokens": 12813281.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 5.325288772583008,
|
|
"epoch": 5.622444343480236,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000261648225796265,
|
|
"loss": 4.7619,
|
|
"mean_token_accuracy": 0.21830219626426697,
|
|
"num_tokens": 12823859.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 5.345687389373779,
|
|
"epoch": 5.626987732848705,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002612954360679499,
|
|
"loss": 4.6506,
|
|
"mean_token_accuracy": 0.22799642831087114,
|
|
"num_tokens": 12833288.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 5.320885181427002,
|
|
"epoch": 5.631531122217174,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000260942680154284,
|
|
"loss": 4.6962,
|
|
"mean_token_accuracy": 0.2263605684041977,
|
|
"num_tokens": 12843698.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 5.293843555450439,
|
|
"epoch": 5.636074511585643,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002605899589256574,
|
|
"loss": 4.6185,
|
|
"mean_token_accuracy": 0.2385869264602661,
|
|
"num_tokens": 12853956.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 5.355259275436401,
|
|
"epoch": 5.640617900954112,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00026023727325237463,
|
|
"loss": 4.7512,
|
|
"mean_token_accuracy": 0.22053562849760056,
|
|
"num_tokens": 12864680.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 5.33907561302185,
|
|
"epoch": 5.645161290322581,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002598846240046527,
|
|
"loss": 4.6606,
|
|
"mean_token_accuracy": 0.2265470504760742,
|
|
"num_tokens": 12873790.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 5.362964344024658,
|
|
"epoch": 5.64970467969105,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002595320120526183,
|
|
"loss": 4.7158,
|
|
"mean_token_accuracy": 0.22927038967609406,
|
|
"num_tokens": 12884559.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 5.306992769241333,
|
|
"epoch": 5.654248069059518,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00025917943826630666,
|
|
"loss": 4.7564,
|
|
"mean_token_accuracy": 0.22505677342414857,
|
|
"num_tokens": 12896176.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 5.357372522354126,
|
|
"epoch": 5.658791458427987,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002588269035156584,
|
|
"loss": 4.7007,
|
|
"mean_token_accuracy": 0.22126404494047164,
|
|
"num_tokens": 12906381.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 5.36885929107666,
|
|
"epoch": 5.663334847796456,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00025847440867051796,
|
|
"loss": 4.6672,
|
|
"mean_token_accuracy": 0.22254393249750137,
|
|
"num_tokens": 12915738.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 5.328704023361206,
|
|
"epoch": 5.667878237164925,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002581219546006314,
|
|
"loss": 4.702,
|
|
"mean_token_accuracy": 0.23178347945213318,
|
|
"num_tokens": 12926059.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 5.312154960632324,
|
|
"epoch": 5.672421626533394,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002577695421756441,
|
|
"loss": 4.7277,
|
|
"mean_token_accuracy": 0.22261458039283752,
|
|
"num_tokens": 12936730.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 5.36267123222351,
|
|
"epoch": 5.676965015901863,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002574171722650986,
|
|
"loss": 4.7232,
|
|
"mean_token_accuracy": 0.22048745006322862,
|
|
"num_tokens": 12945687.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 5.4335943222045895,
|
|
"epoch": 5.681508405270332,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00025706484573843273,
|
|
"loss": 4.7277,
|
|
"mean_token_accuracy": 0.23093274533748626,
|
|
"num_tokens": 12955087.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 5.288049745559692,
|
|
"epoch": 5.6860517946388,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000256712563464977,
|
|
"loss": 4.6931,
|
|
"mean_token_accuracy": 0.22159988582134246,
|
|
"num_tokens": 12965497.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 5.411860513687134,
|
|
"epoch": 5.690595184007269,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002563603263139532,
|
|
"loss": 4.871,
|
|
"mean_token_accuracy": 0.21220864802598954,
|
|
"num_tokens": 12976520.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 5.365349245071411,
|
|
"epoch": 5.695138573375738,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002560081351544711,
|
|
"loss": 4.7088,
|
|
"mean_token_accuracy": 0.227180577814579,
|
|
"num_tokens": 12987271.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 5.35162091255188,
|
|
"epoch": 5.699681962744207,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00025565599085552776,
|
|
"loss": 4.7286,
|
|
"mean_token_accuracy": 0.2198463797569275,
|
|
"num_tokens": 12998170.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 5.297073936462402,
|
|
"epoch": 5.704225352112676,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00025530389428600394,
|
|
"loss": 4.7157,
|
|
"mean_token_accuracy": 0.22597811818122865,
|
|
"num_tokens": 13008059.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 5.335464715957642,
|
|
"epoch": 5.708768741481145,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002549518463146631,
|
|
"loss": 4.6766,
|
|
"mean_token_accuracy": 0.22458689659833908,
|
|
"num_tokens": 13017904.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 5.4298642635345455,
|
|
"epoch": 5.713312130849614,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002545998478101485,
|
|
"loss": 4.7356,
|
|
"mean_token_accuracy": 0.2273765593767166,
|
|
"num_tokens": 13028399.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 5.369762563705445,
|
|
"epoch": 5.717855520218083,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00025424789964098154,
|
|
"loss": 4.7239,
|
|
"mean_token_accuracy": 0.22397335916757583,
|
|
"num_tokens": 13038420.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 5.38324007987976,
|
|
"epoch": 5.722398909586552,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002538960026755593,
|
|
"loss": 4.7361,
|
|
"mean_token_accuracy": 0.22678957730531693,
|
|
"num_tokens": 13048514.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 5.34078016281128,
|
|
"epoch": 5.72694229895502,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00025354415778215255,
|
|
"loss": 4.7864,
|
|
"mean_token_accuracy": 0.22541896253824234,
|
|
"num_tokens": 13059728.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 5.384446859359741,
|
|
"epoch": 5.731485688323489,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00025319236582890364,
|
|
"loss": 4.8008,
|
|
"mean_token_accuracy": 0.2144717365503311,
|
|
"num_tokens": 13069535.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 5.334401369094849,
|
|
"epoch": 5.736029077691958,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00025284062768382424,
|
|
"loss": 4.678,
|
|
"mean_token_accuracy": 0.22169304192066192,
|
|
"num_tokens": 13079012.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 5.3672974586486815,
|
|
"epoch": 5.740572467060427,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002524889442147932,
|
|
"loss": 4.7028,
|
|
"mean_token_accuracy": 0.2366868242621422,
|
|
"num_tokens": 13089067.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 5.338499641418457,
|
|
"epoch": 5.745115856428896,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00025213731628955454,
|
|
"loss": 4.7231,
|
|
"mean_token_accuracy": 0.22626675218343734,
|
|
"num_tokens": 13100576.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 5.380872535705566,
|
|
"epoch": 5.749659245797365,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00025178574477571525,
|
|
"loss": 4.8705,
|
|
"mean_token_accuracy": 0.2118491753935814,
|
|
"num_tokens": 13111926.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 5.316997432708741,
|
|
"epoch": 5.754202635165834,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00025143423054074305,
|
|
"loss": 4.6809,
|
|
"mean_token_accuracy": 0.2224023938179016,
|
|
"num_tokens": 13122362.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 5.3204488277435305,
|
|
"epoch": 5.758746024534303,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002510827744519644,
|
|
"loss": 4.6852,
|
|
"mean_token_accuracy": 0.22675234079360962,
|
|
"num_tokens": 13133372.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 5.328888988494873,
|
|
"epoch": 5.763289413902772,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00025073137737656226,
|
|
"loss": 4.6892,
|
|
"mean_token_accuracy": 0.22804901450872422,
|
|
"num_tokens": 13143440.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 5.37354040145874,
|
|
"epoch": 5.76783280327124,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002503800401815739,
|
|
"loss": 4.7275,
|
|
"mean_token_accuracy": 0.22054435014724733,
|
|
"num_tokens": 13153564.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 5.360758256912232,
|
|
"epoch": 5.772376192639709,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00025002876373388913,
|
|
"loss": 4.8001,
|
|
"mean_token_accuracy": 0.21704500764608384,
|
|
"num_tokens": 13163679.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 5.362074470520019,
|
|
"epoch": 5.776919582008178,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00024967754890024737,
|
|
"loss": 4.7305,
|
|
"mean_token_accuracy": 0.2263677403330803,
|
|
"num_tokens": 13173930.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 5.395290327072144,
|
|
"epoch": 5.781462971376647,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002493263965472366,
|
|
"loss": 4.712,
|
|
"mean_token_accuracy": 0.22807698249816893,
|
|
"num_tokens": 13184805.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 5.33717908859253,
|
|
"epoch": 5.786006360745116,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002489753075412901,
|
|
"loss": 4.7566,
|
|
"mean_token_accuracy": 0.21781749576330184,
|
|
"num_tokens": 13196229.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 5.300772285461425,
|
|
"epoch": 5.790549750113585,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00024862428274868533,
|
|
"loss": 4.644,
|
|
"mean_token_accuracy": 0.23405527472496032,
|
|
"num_tokens": 13205906.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 5.306192827224732,
|
|
"epoch": 5.795093139482054,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002482733230355408,
|
|
"loss": 4.6745,
|
|
"mean_token_accuracy": 0.2229391559958458,
|
|
"num_tokens": 13215076.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 5.412398624420166,
|
|
"epoch": 5.799636528850522,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00024792242926781495,
|
|
"loss": 4.796,
|
|
"mean_token_accuracy": 0.2205319657921791,
|
|
"num_tokens": 13225761.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 5.391673851013183,
|
|
"epoch": 5.804179918218991,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00024757160231130313,
|
|
"loss": 4.7888,
|
|
"mean_token_accuracy": 0.22185304760932922,
|
|
"num_tokens": 13236342.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 5.283697271347046,
|
|
"epoch": 5.80872330758746,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00024722084303163607,
|
|
"loss": 4.6832,
|
|
"mean_token_accuracy": 0.2280586451292038,
|
|
"num_tokens": 13244989.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 5.340116119384765,
|
|
"epoch": 5.813266696955929,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002468701522942774,
|
|
"loss": 4.7062,
|
|
"mean_token_accuracy": 0.22449854016304016,
|
|
"num_tokens": 13255495.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 5.41063494682312,
|
|
"epoch": 5.817810086324398,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00024651953096452165,
|
|
"loss": 4.7691,
|
|
"mean_token_accuracy": 0.22136290222406388,
|
|
"num_tokens": 13266189.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 5.332131385803223,
|
|
"epoch": 5.822353475692867,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00024616897990749196,
|
|
"loss": 4.7162,
|
|
"mean_token_accuracy": 0.22537696212530137,
|
|
"num_tokens": 13276194.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 5.324685668945312,
|
|
"epoch": 5.826896865061336,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002458184999881385,
|
|
"loss": 4.7396,
|
|
"mean_token_accuracy": 0.21797032952308654,
|
|
"num_tokens": 13286188.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 5.365785360336304,
|
|
"epoch": 5.831440254429805,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00024546809207123534,
|
|
"loss": 4.7118,
|
|
"mean_token_accuracy": 0.22861236482858657,
|
|
"num_tokens": 13295926.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 5.3993370056152346,
|
|
"epoch": 5.835983643798274,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002451177570213794,
|
|
"loss": 4.7762,
|
|
"mean_token_accuracy": 0.22297477573156357,
|
|
"num_tokens": 13307373.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 5.325853061676026,
|
|
"epoch": 5.840527033166742,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00024476749570298746,
|
|
"loss": 4.7164,
|
|
"mean_token_accuracy": 0.22135668098926545,
|
|
"num_tokens": 13316984.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 5.316150140762329,
|
|
"epoch": 5.845070422535211,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00024441730898029456,
|
|
"loss": 4.6883,
|
|
"mean_token_accuracy": 0.23036340922117232,
|
|
"num_tokens": 13327188.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 5.468243932723999,
|
|
"epoch": 5.84961381190368,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00024406719771735165,
|
|
"loss": 4.7819,
|
|
"mean_token_accuracy": 0.22192894965410231,
|
|
"num_tokens": 13338108.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 5.328645896911621,
|
|
"epoch": 5.854157201272149,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002437171627780234,
|
|
"loss": 4.6856,
|
|
"mean_token_accuracy": 0.22376096099615098,
|
|
"num_tokens": 13347898.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 5.3635560989379885,
|
|
"epoch": 5.858700590640618,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00024336720502598624,
|
|
"loss": 4.7611,
|
|
"mean_token_accuracy": 0.2251548707485199,
|
|
"num_tokens": 13358649.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 5.3057365894317625,
|
|
"epoch": 5.863243980009087,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002430173253247262,
|
|
"loss": 4.7485,
|
|
"mean_token_accuracy": 0.22342296093702316,
|
|
"num_tokens": 13368447.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 5.31833233833313,
|
|
"epoch": 5.867787369377556,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00024266752453753655,
|
|
"loss": 4.7177,
|
|
"mean_token_accuracy": 0.22020335048437117,
|
|
"num_tokens": 13377809.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 5.387024927139282,
|
|
"epoch": 5.872330758746024,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00024231780352751615,
|
|
"loss": 4.8333,
|
|
"mean_token_accuracy": 0.21462427377700805,
|
|
"num_tokens": 13389352.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 5.391182994842529,
|
|
"epoch": 5.876874148114493,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00024196816315756665,
|
|
"loss": 4.7116,
|
|
"mean_token_accuracy": 0.2249402344226837,
|
|
"num_tokens": 13399929.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 5.374185705184937,
|
|
"epoch": 5.881417537482962,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002416186042903909,
|
|
"loss": 4.6248,
|
|
"mean_token_accuracy": 0.22915414571762086,
|
|
"num_tokens": 13410096.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 5.307699537277221,
|
|
"epoch": 5.885960926851431,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00024126912778849074,
|
|
"loss": 4.8025,
|
|
"mean_token_accuracy": 0.21781497895717622,
|
|
"num_tokens": 13420384.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 5.357935810089112,
|
|
"epoch": 5.8905043162199,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00024091973451416472,
|
|
"loss": 4.7109,
|
|
"mean_token_accuracy": 0.2280397430062294,
|
|
"num_tokens": 13430887.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 5.367037391662597,
|
|
"epoch": 5.895047705588369,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00024057042532950597,
|
|
"loss": 4.7229,
|
|
"mean_token_accuracy": 0.2257548376917839,
|
|
"num_tokens": 13440914.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 5.3338799476623535,
|
|
"epoch": 5.899591094956838,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002402212010964001,
|
|
"loss": 4.7817,
|
|
"mean_token_accuracy": 0.21752348244190217,
|
|
"num_tokens": 13452476.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 5.384309720993042,
|
|
"epoch": 5.904134484325307,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00023987206267652334,
|
|
"loss": 4.699,
|
|
"mean_token_accuracy": 0.22434410750865935,
|
|
"num_tokens": 13463779.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"epoch": 5.904134484325307,
|
|
"eval_entropy": 5.230440338267837,
|
|
"eval_loss": 5.650081157684326,
|
|
"eval_mean_token_accuracy": 0.17452373379884764,
|
|
"eval_num_tokens": 13463779.0,
|
|
"eval_runtime": 1.9924,
|
|
"eval_samples_per_second": 1724.065,
|
|
"eval_steps_per_second": 215.822,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 5.384323406219482,
|
|
"epoch": 5.908677873693776,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00023952301093133993,
|
|
"loss": 4.8153,
|
|
"mean_token_accuracy": 0.21708793491125106,
|
|
"num_tokens": 13475467.0,
|
|
"step": 6505
|
|
},
|
|
{
|
|
"entropy": 5.302642822265625,
|
|
"epoch": 5.913221263062244,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00023917404672210046,
|
|
"loss": 4.6843,
|
|
"mean_token_accuracy": 0.2207771897315979,
|
|
"num_tokens": 13485484.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 5.347629594802856,
|
|
"epoch": 5.917764652430713,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002388251709098392,
|
|
"loss": 4.6942,
|
|
"mean_token_accuracy": 0.22661336064338683,
|
|
"num_tokens": 13494941.0,
|
|
"step": 6515
|
|
},
|
|
{
|
|
"entropy": 5.417224359512329,
|
|
"epoch": 5.922308041799182,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00023847638435537283,
|
|
"loss": 4.7575,
|
|
"mean_token_accuracy": 0.22970338463783263,
|
|
"num_tokens": 13504762.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 5.390254402160645,
|
|
"epoch": 5.926851431167651,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00023812768791929724,
|
|
"loss": 4.7155,
|
|
"mean_token_accuracy": 0.22340993732213973,
|
|
"num_tokens": 13515436.0,
|
|
"step": 6525
|
|
},
|
|
{
|
|
"entropy": 5.354823637008667,
|
|
"epoch": 5.93139482053612,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00023777908246198643,
|
|
"loss": 4.7256,
|
|
"mean_token_accuracy": 0.22847513258457183,
|
|
"num_tokens": 13525211.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 5.333019018173218,
|
|
"epoch": 5.935938209904589,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00023743056884358954,
|
|
"loss": 4.7139,
|
|
"mean_token_accuracy": 0.22193293273448944,
|
|
"num_tokens": 13534954.0,
|
|
"step": 6535
|
|
},
|
|
{
|
|
"entropy": 5.353693532943725,
|
|
"epoch": 5.940481599273058,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023708214792402942,
|
|
"loss": 4.7425,
|
|
"mean_token_accuracy": 0.2274711787700653,
|
|
"num_tokens": 13545459.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 5.309494256973267,
|
|
"epoch": 5.945024988641527,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00023673382056299997,
|
|
"loss": 4.728,
|
|
"mean_token_accuracy": 0.2181580901145935,
|
|
"num_tokens": 13554541.0,
|
|
"step": 6545
|
|
},
|
|
{
|
|
"entropy": 5.373087358474732,
|
|
"epoch": 5.949568378009996,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00023638558761996436,
|
|
"loss": 4.7186,
|
|
"mean_token_accuracy": 0.2223641574382782,
|
|
"num_tokens": 13564810.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 5.344379806518555,
|
|
"epoch": 5.954111767378464,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00023603744995415277,
|
|
"loss": 4.7019,
|
|
"mean_token_accuracy": 0.22599248588085175,
|
|
"num_tokens": 13574756.0,
|
|
"step": 6555
|
|
},
|
|
{
|
|
"entropy": 5.421245813369751,
|
|
"epoch": 5.958655156746933,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00023568940842456026,
|
|
"loss": 4.7604,
|
|
"mean_token_accuracy": 0.2223082110285759,
|
|
"num_tokens": 13584346.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 5.42010531425476,
|
|
"epoch": 5.963198546115402,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002353414638899447,
|
|
"loss": 4.7557,
|
|
"mean_token_accuracy": 0.22410039156675338,
|
|
"num_tokens": 13595713.0,
|
|
"step": 6565
|
|
},
|
|
{
|
|
"entropy": 5.199537706375122,
|
|
"epoch": 5.967741935483871,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023499361720882468,
|
|
"loss": 4.6156,
|
|
"mean_token_accuracy": 0.23457676470279692,
|
|
"num_tokens": 13605065.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 5.352652931213379,
|
|
"epoch": 5.97228532485234,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00023464586923947718,
|
|
"loss": 4.7799,
|
|
"mean_token_accuracy": 0.22116150259971618,
|
|
"num_tokens": 13615028.0,
|
|
"step": 6575
|
|
},
|
|
{
|
|
"entropy": 5.327948331832886,
|
|
"epoch": 5.976828714220809,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00023429822083993596,
|
|
"loss": 4.6983,
|
|
"mean_token_accuracy": 0.22349267899990083,
|
|
"num_tokens": 13625752.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 5.375229024887085,
|
|
"epoch": 5.981372103589278,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00023395067286798872,
|
|
"loss": 4.6748,
|
|
"mean_token_accuracy": 0.23295811265707017,
|
|
"num_tokens": 13635094.0,
|
|
"step": 6585
|
|
},
|
|
{
|
|
"entropy": 5.338564920425415,
|
|
"epoch": 5.985915492957746,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00023360322618117563,
|
|
"loss": 4.7384,
|
|
"mean_token_accuracy": 0.22654549032449722,
|
|
"num_tokens": 13645007.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 5.278239583969116,
|
|
"epoch": 5.990458882326215,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00023325588163678674,
|
|
"loss": 4.6281,
|
|
"mean_token_accuracy": 0.23227672427892684,
|
|
"num_tokens": 13655101.0,
|
|
"step": 6595
|
|
},
|
|
{
|
|
"entropy": 5.315260553359986,
|
|
"epoch": 5.995002271694684,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023290864009186036,
|
|
"loss": 4.6878,
|
|
"mean_token_accuracy": 0.22827746570110322,
|
|
"num_tokens": 13665289.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 5.373373985290527,
|
|
"epoch": 5.999545661063153,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00023256150240318025,
|
|
"loss": 4.781,
|
|
"mean_token_accuracy": 0.22801047265529634,
|
|
"num_tokens": 13675384.0,
|
|
"step": 6605
|
|
},
|
|
{
|
|
"entropy": 5.397659248775906,
|
|
"epoch": 6.003634711494775,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00023221446942727438,
|
|
"loss": 4.6971,
|
|
"mean_token_accuracy": 0.23669642706712088,
|
|
"num_tokens": 13683918.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 5.3271089553833,
|
|
"epoch": 6.008178100863244,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002318675420204119,
|
|
"loss": 4.5972,
|
|
"mean_token_accuracy": 0.23463447242975236,
|
|
"num_tokens": 13693999.0,
|
|
"step": 6615
|
|
},
|
|
{
|
|
"entropy": 5.39841856956482,
|
|
"epoch": 6.012721490231713,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00023152072103860188,
|
|
"loss": 4.6738,
|
|
"mean_token_accuracy": 0.23267799764871597,
|
|
"num_tokens": 13705465.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 5.348963832855224,
|
|
"epoch": 6.017264879600182,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00023117400733759053,
|
|
"loss": 4.5329,
|
|
"mean_token_accuracy": 0.2442344233393669,
|
|
"num_tokens": 13715858.0,
|
|
"step": 6625
|
|
},
|
|
{
|
|
"entropy": 5.365229845046997,
|
|
"epoch": 6.021808268968651,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002308274017728595,
|
|
"loss": 4.5696,
|
|
"mean_token_accuracy": 0.24155094027519225,
|
|
"num_tokens": 13725639.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 5.253775024414063,
|
|
"epoch": 6.026351658337119,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002304809051996235,
|
|
"loss": 4.4803,
|
|
"mean_token_accuracy": 0.24792004823684693,
|
|
"num_tokens": 13736015.0,
|
|
"step": 6635
|
|
},
|
|
{
|
|
"entropy": 5.184731483459473,
|
|
"epoch": 6.030895047705588,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002301345184728284,
|
|
"loss": 4.4508,
|
|
"mean_token_accuracy": 0.24292772114276887,
|
|
"num_tokens": 13745433.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 5.346627855300904,
|
|
"epoch": 6.035438437074057,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00022978824244714903,
|
|
"loss": 4.5588,
|
|
"mean_token_accuracy": 0.2446811765432358,
|
|
"num_tokens": 13754970.0,
|
|
"step": 6645
|
|
},
|
|
{
|
|
"entropy": 5.347137832641602,
|
|
"epoch": 6.039981826442526,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00022944207797698707,
|
|
"loss": 4.6461,
|
|
"mean_token_accuracy": 0.23397691994905473,
|
|
"num_tokens": 13766356.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 5.344094228744507,
|
|
"epoch": 6.044525215810995,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002290960259164689,
|
|
"loss": 4.6013,
|
|
"mean_token_accuracy": 0.24054984599351883,
|
|
"num_tokens": 13776503.0,
|
|
"step": 6655
|
|
},
|
|
{
|
|
"entropy": 5.368874502182007,
|
|
"epoch": 6.049068605179464,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00022875008711944355,
|
|
"loss": 4.6293,
|
|
"mean_token_accuracy": 0.22847393155097961,
|
|
"num_tokens": 13787918.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 5.225936317443848,
|
|
"epoch": 6.053611994547933,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00022840426243948077,
|
|
"loss": 4.5682,
|
|
"mean_token_accuracy": 0.2391144573688507,
|
|
"num_tokens": 13798305.0,
|
|
"step": 6665
|
|
},
|
|
{
|
|
"entropy": 5.340609121322632,
|
|
"epoch": 6.058155383916402,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00022805855272986842,
|
|
"loss": 4.5792,
|
|
"mean_token_accuracy": 0.23504221737384795,
|
|
"num_tokens": 13808867.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 5.309413909912109,
|
|
"epoch": 6.062698773284871,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00022771295884361088,
|
|
"loss": 4.4616,
|
|
"mean_token_accuracy": 0.24842358231544495,
|
|
"num_tokens": 13820073.0,
|
|
"step": 6675
|
|
},
|
|
{
|
|
"entropy": 5.356910228729248,
|
|
"epoch": 6.067242162653339,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002273674816334268,
|
|
"loss": 4.6241,
|
|
"mean_token_accuracy": 0.23853846192359923,
|
|
"num_tokens": 13830318.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 5.214842033386231,
|
|
"epoch": 6.071785552021808,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022702212195174664,
|
|
"loss": 4.5572,
|
|
"mean_token_accuracy": 0.23498122096061708,
|
|
"num_tokens": 13840847.0,
|
|
"step": 6685
|
|
},
|
|
{
|
|
"entropy": 5.32514500617981,
|
|
"epoch": 6.076328941390277,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00022667688065071146,
|
|
"loss": 4.5645,
|
|
"mean_token_accuracy": 0.24000952541828155,
|
|
"num_tokens": 13850478.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 5.338671875,
|
|
"epoch": 6.080872330758746,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00022633175858216936,
|
|
"loss": 4.6251,
|
|
"mean_token_accuracy": 0.23380353748798371,
|
|
"num_tokens": 13860749.0,
|
|
"step": 6695
|
|
},
|
|
{
|
|
"entropy": 5.351956605911255,
|
|
"epoch": 6.085415720127215,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00022598675659767515,
|
|
"loss": 4.6085,
|
|
"mean_token_accuracy": 0.23572426289319992,
|
|
"num_tokens": 13869909.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 5.323867702484131,
|
|
"epoch": 6.089959109495684,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00022564187554848663,
|
|
"loss": 4.5543,
|
|
"mean_token_accuracy": 0.2430637776851654,
|
|
"num_tokens": 13880235.0,
|
|
"step": 6705
|
|
},
|
|
{
|
|
"entropy": 5.203256893157959,
|
|
"epoch": 6.094502498864153,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00022529711628556387,
|
|
"loss": 4.4797,
|
|
"mean_token_accuracy": 0.24931784272193908,
|
|
"num_tokens": 13890324.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 5.346428108215332,
|
|
"epoch": 6.099045888232622,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00022495247965956567,
|
|
"loss": 4.5985,
|
|
"mean_token_accuracy": 0.23465524166822432,
|
|
"num_tokens": 13900250.0,
|
|
"step": 6715
|
|
},
|
|
{
|
|
"entropy": 5.339940071105957,
|
|
"epoch": 6.10358927760109,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002246079665208489,
|
|
"loss": 4.559,
|
|
"mean_token_accuracy": 0.22917823493480682,
|
|
"num_tokens": 13909736.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 5.29451093673706,
|
|
"epoch": 6.108132666969559,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00022426357771946544,
|
|
"loss": 4.5831,
|
|
"mean_token_accuracy": 0.24051479995250702,
|
|
"num_tokens": 13920187.0,
|
|
"step": 6725
|
|
},
|
|
{
|
|
"entropy": 5.308684062957764,
|
|
"epoch": 6.112676056338028,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00022391931410516036,
|
|
"loss": 4.5335,
|
|
"mean_token_accuracy": 0.23218681812286376,
|
|
"num_tokens": 13930858.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 5.2485997676849365,
|
|
"epoch": 6.117219445706497,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00022357517652736992,
|
|
"loss": 4.5016,
|
|
"mean_token_accuracy": 0.25332304239273074,
|
|
"num_tokens": 13940771.0,
|
|
"step": 6735
|
|
},
|
|
{
|
|
"entropy": 5.313397932052612,
|
|
"epoch": 6.121762835074966,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00022323116583521948,
|
|
"loss": 4.645,
|
|
"mean_token_accuracy": 0.22977212965488433,
|
|
"num_tokens": 13951312.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 5.343466758728027,
|
|
"epoch": 6.126306224443435,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00022288728287752109,
|
|
"loss": 4.6068,
|
|
"mean_token_accuracy": 0.23486659526824952,
|
|
"num_tokens": 13960862.0,
|
|
"step": 6745
|
|
},
|
|
{
|
|
"entropy": 5.356136131286621,
|
|
"epoch": 6.130849613811904,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002225435285027719,
|
|
"loss": 4.5639,
|
|
"mean_token_accuracy": 0.24102093279361725,
|
|
"num_tokens": 13972228.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 5.273715209960938,
|
|
"epoch": 6.135393003180373,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00022219990355915148,
|
|
"loss": 4.5555,
|
|
"mean_token_accuracy": 0.23981836587190627,
|
|
"num_tokens": 13982977.0,
|
|
"step": 6755
|
|
},
|
|
{
|
|
"entropy": 5.3297261714935305,
|
|
"epoch": 6.139936392548841,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00022185640889452036,
|
|
"loss": 4.5339,
|
|
"mean_token_accuracy": 0.2457814857363701,
|
|
"num_tokens": 13991645.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 5.340643978118896,
|
|
"epoch": 6.14447978191731,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00022151304535641747,
|
|
"loss": 4.5707,
|
|
"mean_token_accuracy": 0.23557541370391846,
|
|
"num_tokens": 14001313.0,
|
|
"step": 6765
|
|
},
|
|
{
|
|
"entropy": 5.310829448699951,
|
|
"epoch": 6.149023171285779,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002211698137920582,
|
|
"loss": 4.6123,
|
|
"mean_token_accuracy": 0.23521765619516372,
|
|
"num_tokens": 14011248.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 5.325975465774536,
|
|
"epoch": 6.153566560654248,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00022082671504833223,
|
|
"loss": 4.5707,
|
|
"mean_token_accuracy": 0.23310375660657884,
|
|
"num_tokens": 14022290.0,
|
|
"step": 6775
|
|
},
|
|
{
|
|
"entropy": 5.249567174911499,
|
|
"epoch": 6.158109950022717,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00022048374997180175,
|
|
"loss": 4.555,
|
|
"mean_token_accuracy": 0.24026138782501222,
|
|
"num_tokens": 14032899.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 5.330997180938721,
|
|
"epoch": 6.162653339391186,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00022014091940869885,
|
|
"loss": 4.5849,
|
|
"mean_token_accuracy": 0.2426824152469635,
|
|
"num_tokens": 14043065.0,
|
|
"step": 6785
|
|
},
|
|
{
|
|
"entropy": 5.309734153747558,
|
|
"epoch": 6.167196728759655,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00021979822420492402,
|
|
"loss": 4.6253,
|
|
"mean_token_accuracy": 0.23417572528123856,
|
|
"num_tokens": 14054594.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 5.449472665786743,
|
|
"epoch": 6.171740118128124,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00021945566520604337,
|
|
"loss": 4.6846,
|
|
"mean_token_accuracy": 0.23080552369356155,
|
|
"num_tokens": 14064211.0,
|
|
"step": 6795
|
|
},
|
|
{
|
|
"entropy": 5.314081907272339,
|
|
"epoch": 6.176283507496592,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002191132432572875,
|
|
"loss": 4.568,
|
|
"mean_token_accuracy": 0.2450114443898201,
|
|
"num_tokens": 14073946.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 5.303177738189698,
|
|
"epoch": 6.180826896865061,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00021877095920354818,
|
|
"loss": 4.5066,
|
|
"mean_token_accuracy": 0.24394751638174056,
|
|
"num_tokens": 14083303.0,
|
|
"step": 6805
|
|
},
|
|
{
|
|
"entropy": 5.319565916061402,
|
|
"epoch": 6.18537028623353,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00021842881388937764,
|
|
"loss": 4.5011,
|
|
"mean_token_accuracy": 0.25176720321178436,
|
|
"num_tokens": 14093475.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 5.360175657272339,
|
|
"epoch": 6.189913675601999,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00021808680815898508,
|
|
"loss": 4.6511,
|
|
"mean_token_accuracy": 0.22806433886289595,
|
|
"num_tokens": 14104117.0,
|
|
"step": 6815
|
|
},
|
|
{
|
|
"entropy": 5.258196926116943,
|
|
"epoch": 6.194457064970468,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00021774494285623598,
|
|
"loss": 4.5234,
|
|
"mean_token_accuracy": 0.24171907156705857,
|
|
"num_tokens": 14115480.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 5.22868275642395,
|
|
"epoch": 6.199000454338937,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002174032188246487,
|
|
"loss": 4.5354,
|
|
"mean_token_accuracy": 0.2360183358192444,
|
|
"num_tokens": 14125827.0,
|
|
"step": 6825
|
|
},
|
|
{
|
|
"entropy": 5.323046922683716,
|
|
"epoch": 6.203543843707406,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00021706163690739344,
|
|
"loss": 4.5676,
|
|
"mean_token_accuracy": 0.24473971277475357,
|
|
"num_tokens": 14136424.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 5.365231943130493,
|
|
"epoch": 6.208087233075875,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00021672019794728963,
|
|
"loss": 4.6774,
|
|
"mean_token_accuracy": 0.2276274412870407,
|
|
"num_tokens": 14147478.0,
|
|
"step": 6835
|
|
},
|
|
{
|
|
"entropy": 5.381686973571777,
|
|
"epoch": 6.212630622444343,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00021637890278680406,
|
|
"loss": 4.654,
|
|
"mean_token_accuracy": 0.23763065189123153,
|
|
"num_tokens": 14159044.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 5.3135429382324215,
|
|
"epoch": 6.217174011812812,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00021603775226804845,
|
|
"loss": 4.5819,
|
|
"mean_token_accuracy": 0.23911372125148772,
|
|
"num_tokens": 14169370.0,
|
|
"step": 6845
|
|
},
|
|
{
|
|
"entropy": 5.333434104919434,
|
|
"epoch": 6.221717401181281,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00021569674723277783,
|
|
"loss": 4.6205,
|
|
"mean_token_accuracy": 0.23191130459308623,
|
|
"num_tokens": 14179774.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 5.327278566360474,
|
|
"epoch": 6.22626079054975,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00021535588852238823,
|
|
"loss": 4.5672,
|
|
"mean_token_accuracy": 0.2461705282330513,
|
|
"num_tokens": 14190182.0,
|
|
"step": 6855
|
|
},
|
|
{
|
|
"entropy": 5.266404247283935,
|
|
"epoch": 6.230804179918219,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00021501517697791479,
|
|
"loss": 4.491,
|
|
"mean_token_accuracy": 0.24954034090042115,
|
|
"num_tokens": 14199505.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 5.3499298095703125,
|
|
"epoch": 6.235347569286688,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002146746134400291,
|
|
"loss": 4.6209,
|
|
"mean_token_accuracy": 0.22750377058982849,
|
|
"num_tokens": 14209677.0,
|
|
"step": 6865
|
|
},
|
|
{
|
|
"entropy": 5.284366035461426,
|
|
"epoch": 6.239890958655157,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00021433419874903803,
|
|
"loss": 4.5437,
|
|
"mean_token_accuracy": 0.23810020089149475,
|
|
"num_tokens": 14219356.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 5.227415275573731,
|
|
"epoch": 6.244434348023626,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00021399393374488097,
|
|
"loss": 4.4993,
|
|
"mean_token_accuracy": 0.2501468524336815,
|
|
"num_tokens": 14227992.0,
|
|
"step": 6875
|
|
},
|
|
{
|
|
"entropy": 5.3388283252716064,
|
|
"epoch": 6.248977737392094,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00021365381926712797,
|
|
"loss": 4.5638,
|
|
"mean_token_accuracy": 0.24253699332475662,
|
|
"num_tokens": 14238354.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 5.331574726104736,
|
|
"epoch": 6.253521126760563,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00021331385615497756,
|
|
"loss": 4.5981,
|
|
"mean_token_accuracy": 0.23565790355205535,
|
|
"num_tokens": 14249560.0,
|
|
"step": 6885
|
|
},
|
|
{
|
|
"entropy": 5.297011184692383,
|
|
"epoch": 6.258064516129032,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002129740452472552,
|
|
"loss": 4.6094,
|
|
"mean_token_accuracy": 0.23390067964792252,
|
|
"num_tokens": 14259372.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 5.299483776092529,
|
|
"epoch": 6.262607905497501,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00021263438738241026,
|
|
"loss": 4.5401,
|
|
"mean_token_accuracy": 0.23972143977880478,
|
|
"num_tokens": 14270837.0,
|
|
"step": 6895
|
|
},
|
|
{
|
|
"entropy": 5.320937252044677,
|
|
"epoch": 6.26715129486597,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00021229488339851498,
|
|
"loss": 4.5433,
|
|
"mean_token_accuracy": 0.23749537616968155,
|
|
"num_tokens": 14280138.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 5.34907717704773,
|
|
"epoch": 6.271694684234439,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00021195553413326146,
|
|
"loss": 4.6634,
|
|
"mean_token_accuracy": 0.2326961040496826,
|
|
"num_tokens": 14290131.0,
|
|
"step": 6905
|
|
},
|
|
{
|
|
"entropy": 5.273083829879761,
|
|
"epoch": 6.276238073602908,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002116163404239605,
|
|
"loss": 4.5888,
|
|
"mean_token_accuracy": 0.23499025106430055,
|
|
"num_tokens": 14299961.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 5.265652179718018,
|
|
"epoch": 6.280781462971377,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021127730310753866,
|
|
"loss": 4.5442,
|
|
"mean_token_accuracy": 0.23856717497110366,
|
|
"num_tokens": 14310529.0,
|
|
"step": 6915
|
|
},
|
|
{
|
|
"entropy": 5.274373865127563,
|
|
"epoch": 6.285324852339846,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00021093842302053706,
|
|
"loss": 4.4891,
|
|
"mean_token_accuracy": 0.2522676795721054,
|
|
"num_tokens": 14320220.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 5.40164999961853,
|
|
"epoch": 6.289868241708314,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00021059970099910835,
|
|
"loss": 4.6227,
|
|
"mean_token_accuracy": 0.2374628409743309,
|
|
"num_tokens": 14331700.0,
|
|
"step": 6925
|
|
},
|
|
{
|
|
"entropy": 5.308527898788452,
|
|
"epoch": 6.294411631076783,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00021026113787901568,
|
|
"loss": 4.5517,
|
|
"mean_token_accuracy": 0.23574525415897368,
|
|
"num_tokens": 14341163.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 5.2594075202941895,
|
|
"epoch": 6.298955020445252,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00020992273449562966,
|
|
"loss": 4.5681,
|
|
"mean_token_accuracy": 0.24165342152118682,
|
|
"num_tokens": 14351398.0,
|
|
"step": 6935
|
|
},
|
|
{
|
|
"entropy": 5.349909925460816,
|
|
"epoch": 6.303498409813721,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00020958449168392723,
|
|
"loss": 4.5819,
|
|
"mean_token_accuracy": 0.23733491599559783,
|
|
"num_tokens": 14361920.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 5.281101179122925,
|
|
"epoch": 6.30804179918219,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002092464102784888,
|
|
"loss": 4.5694,
|
|
"mean_token_accuracy": 0.2427244871854782,
|
|
"num_tokens": 14371980.0,
|
|
"step": 6945
|
|
},
|
|
{
|
|
"entropy": 5.310888051986694,
|
|
"epoch": 6.312585188550659,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00020890849111349663,
|
|
"loss": 4.5831,
|
|
"mean_token_accuracy": 0.23156099319458007,
|
|
"num_tokens": 14381397.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 5.281076717376709,
|
|
"epoch": 6.317128577919128,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00020857073502273266,
|
|
"loss": 4.6049,
|
|
"mean_token_accuracy": 0.2329518660902977,
|
|
"num_tokens": 14393476.0,
|
|
"step": 6955
|
|
},
|
|
{
|
|
"entropy": 5.326323413848877,
|
|
"epoch": 6.321671967287597,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002082331428395765,
|
|
"loss": 4.6141,
|
|
"mean_token_accuracy": 0.23737093210220336,
|
|
"num_tokens": 14404801.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 5.33033595085144,
|
|
"epoch": 6.326215356656065,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00020789571539700326,
|
|
"loss": 4.5988,
|
|
"mean_token_accuracy": 0.23399335145950317,
|
|
"num_tokens": 14415921.0,
|
|
"step": 6965
|
|
},
|
|
{
|
|
"entropy": 5.27001919746399,
|
|
"epoch": 6.330758746024534,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020755845352758175,
|
|
"loss": 4.6356,
|
|
"mean_token_accuracy": 0.23417109549045562,
|
|
"num_tokens": 14426455.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 5.3818565845489506,
|
|
"epoch": 6.335302135393003,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002072213580634718,
|
|
"loss": 4.5962,
|
|
"mean_token_accuracy": 0.23974624574184417,
|
|
"num_tokens": 14438720.0,
|
|
"step": 6975
|
|
},
|
|
{
|
|
"entropy": 5.323555421829224,
|
|
"epoch": 6.339845524761472,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002068844298364233,
|
|
"loss": 4.6178,
|
|
"mean_token_accuracy": 0.23464094698429108,
|
|
"num_tokens": 14449644.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 5.308932542800903,
|
|
"epoch": 6.344388914129941,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00020654766967777288,
|
|
"loss": 4.6355,
|
|
"mean_token_accuracy": 0.22669494301080703,
|
|
"num_tokens": 14459712.0,
|
|
"step": 6985
|
|
},
|
|
{
|
|
"entropy": 5.30234842300415,
|
|
"epoch": 6.34893230349841,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002062110784184431,
|
|
"loss": 4.5948,
|
|
"mean_token_accuracy": 0.23719390630722045,
|
|
"num_tokens": 14469440.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 5.326094913482666,
|
|
"epoch": 6.353475692866879,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00020587465688893904,
|
|
"loss": 4.6392,
|
|
"mean_token_accuracy": 0.2229648619890213,
|
|
"num_tokens": 14480587.0,
|
|
"step": 6995
|
|
},
|
|
{
|
|
"entropy": 5.403822755813598,
|
|
"epoch": 6.358019082235348,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00020553840591934765,
|
|
"loss": 4.652,
|
|
"mean_token_accuracy": 0.23423184007406234,
|
|
"num_tokens": 14490360.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"epoch": 6.358019082235348,
|
|
"eval_entropy": 5.223701049006262,
|
|
"eval_loss": 5.665815830230713,
|
|
"eval_mean_token_accuracy": 0.17393263801932335,
|
|
"eval_num_tokens": 14490360.0,
|
|
"eval_runtime": 1.9979,
|
|
"eval_samples_per_second": 1719.336,
|
|
"eval_steps_per_second": 215.23,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"entropy": 5.319053697586059,
|
|
"epoch": 6.362562471603816,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002052023263393346,
|
|
"loss": 4.5726,
|
|
"mean_token_accuracy": 0.23670225739479064,
|
|
"num_tokens": 14501135.0,
|
|
"step": 7005
|
|
},
|
|
{
|
|
"entropy": 5.331496381759644,
|
|
"epoch": 6.367105860972285,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00020486641897814295,
|
|
"loss": 4.5958,
|
|
"mean_token_accuracy": 0.23614376187324523,
|
|
"num_tokens": 14511828.0,
|
|
"step": 7010
|
|
},
|
|
{
|
|
"entropy": 5.293062734603882,
|
|
"epoch": 6.371649250340754,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00020453068466459067,
|
|
"loss": 4.584,
|
|
"mean_token_accuracy": 0.2360141322016716,
|
|
"num_tokens": 14521973.0,
|
|
"step": 7015
|
|
},
|
|
{
|
|
"entropy": 5.370773458480835,
|
|
"epoch": 6.376192639709223,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00020419512422706893,
|
|
"loss": 4.6289,
|
|
"mean_token_accuracy": 0.24203862249851227,
|
|
"num_tokens": 14532515.0,
|
|
"step": 7020
|
|
},
|
|
{
|
|
"entropy": 5.315709924697876,
|
|
"epoch": 6.380736029077692,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00020385973849353957,
|
|
"loss": 4.5739,
|
|
"mean_token_accuracy": 0.2353421077132225,
|
|
"num_tokens": 14542577.0,
|
|
"step": 7025
|
|
},
|
|
{
|
|
"entropy": 5.269266128540039,
|
|
"epoch": 6.385279418446161,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00020352452829153369,
|
|
"loss": 4.4928,
|
|
"mean_token_accuracy": 0.24236989468336106,
|
|
"num_tokens": 14552472.0,
|
|
"step": 7030
|
|
},
|
|
{
|
|
"entropy": 5.287693500518799,
|
|
"epoch": 6.38982280781463,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000203189494448149,
|
|
"loss": 4.5472,
|
|
"mean_token_accuracy": 0.24272410422563553,
|
|
"num_tokens": 14562615.0,
|
|
"step": 7035
|
|
},
|
|
{
|
|
"entropy": 5.272085380554199,
|
|
"epoch": 6.394366197183099,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00020285463779004837,
|
|
"loss": 4.574,
|
|
"mean_token_accuracy": 0.24059284776449202,
|
|
"num_tokens": 14573200.0,
|
|
"step": 7040
|
|
},
|
|
{
|
|
"entropy": 5.371761417388916,
|
|
"epoch": 6.398909586551568,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020251995914345722,
|
|
"loss": 4.593,
|
|
"mean_token_accuracy": 0.24257361888885498,
|
|
"num_tokens": 14583042.0,
|
|
"step": 7045
|
|
},
|
|
{
|
|
"entropy": 5.407893419265747,
|
|
"epoch": 6.403452975920036,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00020218545933416181,
|
|
"loss": 4.6359,
|
|
"mean_token_accuracy": 0.2358793705701828,
|
|
"num_tokens": 14594409.0,
|
|
"step": 7050
|
|
},
|
|
{
|
|
"entropy": 5.357564306259155,
|
|
"epoch": 6.407996365288505,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00020185113918750724,
|
|
"loss": 4.6842,
|
|
"mean_token_accuracy": 0.2315233662724495,
|
|
"num_tokens": 14605370.0,
|
|
"step": 7055
|
|
},
|
|
{
|
|
"entropy": 5.3325706958770756,
|
|
"epoch": 6.412539754656974,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002015169995283952,
|
|
"loss": 4.6062,
|
|
"mean_token_accuracy": 0.2364664152264595,
|
|
"num_tokens": 14615056.0,
|
|
"step": 7060
|
|
},
|
|
{
|
|
"entropy": 5.4302330493927,
|
|
"epoch": 6.417083144025443,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00020118304118128211,
|
|
"loss": 4.683,
|
|
"mean_token_accuracy": 0.2255544051527977,
|
|
"num_tokens": 14624959.0,
|
|
"step": 7065
|
|
},
|
|
{
|
|
"entropy": 5.2867734909057615,
|
|
"epoch": 6.421626533393912,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00020084926497017703,
|
|
"loss": 4.5401,
|
|
"mean_token_accuracy": 0.2428005114197731,
|
|
"num_tokens": 14635348.0,
|
|
"step": 7070
|
|
},
|
|
{
|
|
"entropy": 5.267073106765747,
|
|
"epoch": 6.426169922762381,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00020051567171863944,
|
|
"loss": 4.6036,
|
|
"mean_token_accuracy": 0.23428304195404054,
|
|
"num_tokens": 14645232.0,
|
|
"step": 7075
|
|
},
|
|
{
|
|
"entropy": 5.313785696029663,
|
|
"epoch": 6.43071331213085,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002001822622497777,
|
|
"loss": 4.5753,
|
|
"mean_token_accuracy": 0.24492418318986892,
|
|
"num_tokens": 14656194.0,
|
|
"step": 7080
|
|
},
|
|
{
|
|
"entropy": 5.351431512832642,
|
|
"epoch": 6.435256701499318,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00019984903738624632,
|
|
"loss": 4.6086,
|
|
"mean_token_accuracy": 0.23863725364208221,
|
|
"num_tokens": 14667019.0,
|
|
"step": 7085
|
|
},
|
|
{
|
|
"entropy": 5.354405164718628,
|
|
"epoch": 6.439800090867787,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00019951599795024478,
|
|
"loss": 4.5898,
|
|
"mean_token_accuracy": 0.2405167981982231,
|
|
"num_tokens": 14676339.0,
|
|
"step": 7090
|
|
},
|
|
{
|
|
"entropy": 5.348089075088501,
|
|
"epoch": 6.444343480236256,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00019918314476351462,
|
|
"loss": 4.6153,
|
|
"mean_token_accuracy": 0.23513559103012086,
|
|
"num_tokens": 14688206.0,
|
|
"step": 7095
|
|
},
|
|
{
|
|
"entropy": 5.29837818145752,
|
|
"epoch": 6.448886869604725,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00019885047864733803,
|
|
"loss": 4.5653,
|
|
"mean_token_accuracy": 0.24184782952070236,
|
|
"num_tokens": 14698397.0,
|
|
"step": 7100
|
|
},
|
|
{
|
|
"entropy": 5.314942502975464,
|
|
"epoch": 6.453430258973194,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00019851800042253553,
|
|
"loss": 4.5346,
|
|
"mean_token_accuracy": 0.2434400960803032,
|
|
"num_tokens": 14708140.0,
|
|
"step": 7105
|
|
},
|
|
{
|
|
"entropy": 5.321618843078613,
|
|
"epoch": 6.457973648341663,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001981857109094642,
|
|
"loss": 4.5924,
|
|
"mean_token_accuracy": 0.2412719264626503,
|
|
"num_tokens": 14718710.0,
|
|
"step": 7110
|
|
},
|
|
{
|
|
"entropy": 5.318980979919433,
|
|
"epoch": 6.462517037710132,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00019785361092801524,
|
|
"loss": 4.6057,
|
|
"mean_token_accuracy": 0.23292847126722335,
|
|
"num_tokens": 14729143.0,
|
|
"step": 7115
|
|
},
|
|
{
|
|
"entropy": 5.317721366882324,
|
|
"epoch": 6.467060427078601,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00019752170129761244,
|
|
"loss": 4.5804,
|
|
"mean_token_accuracy": 0.24303565472364425,
|
|
"num_tokens": 14739386.0,
|
|
"step": 7120
|
|
},
|
|
{
|
|
"entropy": 5.312438154220581,
|
|
"epoch": 6.47160381644707,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001971899828372098,
|
|
"loss": 4.6166,
|
|
"mean_token_accuracy": 0.23563313037157058,
|
|
"num_tokens": 14749209.0,
|
|
"step": 7125
|
|
},
|
|
{
|
|
"entropy": 5.3570434093475345,
|
|
"epoch": 6.476147205815538,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00019685845636528972,
|
|
"loss": 4.6196,
|
|
"mean_token_accuracy": 0.24506738781929016,
|
|
"num_tokens": 14759351.0,
|
|
"step": 7130
|
|
},
|
|
{
|
|
"entropy": 5.375803089141845,
|
|
"epoch": 6.480690595184007,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00019652712269986074,
|
|
"loss": 4.5951,
|
|
"mean_token_accuracy": 0.23841550648212434,
|
|
"num_tokens": 14769069.0,
|
|
"step": 7135
|
|
},
|
|
{
|
|
"entropy": 5.375771856307983,
|
|
"epoch": 6.485233984552476,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00019619598265845556,
|
|
"loss": 4.6002,
|
|
"mean_token_accuracy": 0.2401449829339981,
|
|
"num_tokens": 14778563.0,
|
|
"step": 7140
|
|
},
|
|
{
|
|
"entropy": 5.258846187591553,
|
|
"epoch": 6.489777373920945,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00019586503705812969,
|
|
"loss": 4.5741,
|
|
"mean_token_accuracy": 0.24175891578197478,
|
|
"num_tokens": 14788305.0,
|
|
"step": 7145
|
|
},
|
|
{
|
|
"entropy": 5.336733675003051,
|
|
"epoch": 6.494320763289414,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001955342867154582,
|
|
"loss": 4.6442,
|
|
"mean_token_accuracy": 0.23137862533330916,
|
|
"num_tokens": 14799463.0,
|
|
"step": 7150
|
|
},
|
|
{
|
|
"entropy": 5.2802564144134525,
|
|
"epoch": 6.498864152657883,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00019520373244653484,
|
|
"loss": 4.5454,
|
|
"mean_token_accuracy": 0.24898803383111953,
|
|
"num_tokens": 14809516.0,
|
|
"step": 7155
|
|
},
|
|
{
|
|
"entropy": 5.299689245223999,
|
|
"epoch": 6.503407542026352,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001948733750669693,
|
|
"loss": 4.6537,
|
|
"mean_token_accuracy": 0.22857050895690917,
|
|
"num_tokens": 14819605.0,
|
|
"step": 7160
|
|
},
|
|
{
|
|
"entropy": 5.3258056640625,
|
|
"epoch": 6.50795093139482,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00019454321539188568,
|
|
"loss": 4.6013,
|
|
"mean_token_accuracy": 0.2334933027625084,
|
|
"num_tokens": 14830241.0,
|
|
"step": 7165
|
|
},
|
|
{
|
|
"entropy": 5.35510573387146,
|
|
"epoch": 6.512494320763289,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00019421325423592007,
|
|
"loss": 4.6152,
|
|
"mean_token_accuracy": 0.24001964926719666,
|
|
"num_tokens": 14840294.0,
|
|
"step": 7170
|
|
},
|
|
{
|
|
"entropy": 5.366328287124634,
|
|
"epoch": 6.517037710131758,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001938834924132189,
|
|
"loss": 4.6449,
|
|
"mean_token_accuracy": 0.23424689322710038,
|
|
"num_tokens": 14850156.0,
|
|
"step": 7175
|
|
},
|
|
{
|
|
"entropy": 5.285260581970215,
|
|
"epoch": 6.521581099500227,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001935539307374366,
|
|
"loss": 4.5397,
|
|
"mean_token_accuracy": 0.24958601742982864,
|
|
"num_tokens": 14859889.0,
|
|
"step": 7180
|
|
},
|
|
{
|
|
"entropy": 5.34511981010437,
|
|
"epoch": 6.526124488868696,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00019322457002173395,
|
|
"loss": 4.6158,
|
|
"mean_token_accuracy": 0.24052239656448365,
|
|
"num_tokens": 14871659.0,
|
|
"step": 7185
|
|
},
|
|
{
|
|
"entropy": 5.351465368270874,
|
|
"epoch": 6.530667878237165,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00019289541107877562,
|
|
"loss": 4.5938,
|
|
"mean_token_accuracy": 0.24129276871681213,
|
|
"num_tokens": 14882041.0,
|
|
"step": 7190
|
|
},
|
|
{
|
|
"entropy": 5.320094633102417,
|
|
"epoch": 6.535211267605634,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00019256645472072876,
|
|
"loss": 4.5661,
|
|
"mean_token_accuracy": 0.243361958861351,
|
|
"num_tokens": 14891709.0,
|
|
"step": 7195
|
|
},
|
|
{
|
|
"entropy": 5.321542882919312,
|
|
"epoch": 6.539754656974103,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00019223770175926032,
|
|
"loss": 4.6014,
|
|
"mean_token_accuracy": 0.23953818380832673,
|
|
"num_tokens": 14901898.0,
|
|
"step": 7200
|
|
},
|
|
{
|
|
"entropy": 5.26777286529541,
|
|
"epoch": 6.544298046342572,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00019190915300553574,
|
|
"loss": 4.525,
|
|
"mean_token_accuracy": 0.2449884757399559,
|
|
"num_tokens": 14911910.0,
|
|
"step": 7205
|
|
},
|
|
{
|
|
"entropy": 5.29623851776123,
|
|
"epoch": 6.54884143571104,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001915808092702162,
|
|
"loss": 4.567,
|
|
"mean_token_accuracy": 0.23504744470119476,
|
|
"num_tokens": 14922055.0,
|
|
"step": 7210
|
|
},
|
|
{
|
|
"entropy": 5.2830089092254635,
|
|
"epoch": 6.553384825079509,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001912526713634574,
|
|
"loss": 4.5478,
|
|
"mean_token_accuracy": 0.24724195301532745,
|
|
"num_tokens": 14931841.0,
|
|
"step": 7215
|
|
},
|
|
{
|
|
"entropy": 5.2922447681427,
|
|
"epoch": 6.557928214447978,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001909247400949069,
|
|
"loss": 4.5722,
|
|
"mean_token_accuracy": 0.24079734683036805,
|
|
"num_tokens": 14942356.0,
|
|
"step": 7220
|
|
},
|
|
{
|
|
"entropy": 5.288787746429444,
|
|
"epoch": 6.562471603816447,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00019059701627370263,
|
|
"loss": 4.5575,
|
|
"mean_token_accuracy": 0.2390957146883011,
|
|
"num_tokens": 14952552.0,
|
|
"step": 7225
|
|
},
|
|
{
|
|
"entropy": 5.258346462249756,
|
|
"epoch": 6.567014993184916,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00019026950070847025,
|
|
"loss": 4.5729,
|
|
"mean_token_accuracy": 0.23907023966312407,
|
|
"num_tokens": 14962728.0,
|
|
"step": 7230
|
|
},
|
|
{
|
|
"entropy": 5.254478454589844,
|
|
"epoch": 6.571558382553385,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00018994219420732228,
|
|
"loss": 4.5434,
|
|
"mean_token_accuracy": 0.2482568606734276,
|
|
"num_tokens": 14974071.0,
|
|
"step": 7235
|
|
},
|
|
{
|
|
"entropy": 5.340001153945923,
|
|
"epoch": 6.576101771921854,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00018961509757785462,
|
|
"loss": 4.6494,
|
|
"mean_token_accuracy": 0.23797639161348344,
|
|
"num_tokens": 14983567.0,
|
|
"step": 7240
|
|
},
|
|
{
|
|
"entropy": 5.274724531173706,
|
|
"epoch": 6.580645161290323,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00018928821162714587,
|
|
"loss": 4.5675,
|
|
"mean_token_accuracy": 0.24615298062562943,
|
|
"num_tokens": 14994172.0,
|
|
"step": 7245
|
|
},
|
|
{
|
|
"entropy": 5.308278751373291,
|
|
"epoch": 6.585188550658792,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00018896153716175458,
|
|
"loss": 4.5516,
|
|
"mean_token_accuracy": 0.2419088065624237,
|
|
"num_tokens": 15004094.0,
|
|
"step": 7250
|
|
},
|
|
{
|
|
"entropy": 5.292300415039063,
|
|
"epoch": 6.58973194002726,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001886350749877175,
|
|
"loss": 4.5733,
|
|
"mean_token_accuracy": 0.23700589388608934,
|
|
"num_tokens": 15013986.0,
|
|
"step": 7255
|
|
},
|
|
{
|
|
"entropy": 5.304951000213623,
|
|
"epoch": 6.594275329395729,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001883088259105476,
|
|
"loss": 4.5989,
|
|
"mean_token_accuracy": 0.23697343319654465,
|
|
"num_tokens": 15024764.0,
|
|
"step": 7260
|
|
},
|
|
{
|
|
"entropy": 5.346390628814698,
|
|
"epoch": 6.598818718764198,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00018798279073523204,
|
|
"loss": 4.6208,
|
|
"mean_token_accuracy": 0.2322131648659706,
|
|
"num_tokens": 15034971.0,
|
|
"step": 7265
|
|
},
|
|
{
|
|
"entropy": 5.33525333404541,
|
|
"epoch": 6.603362108132667,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00018765697026623017,
|
|
"loss": 4.6779,
|
|
"mean_token_accuracy": 0.2258239910006523,
|
|
"num_tokens": 15045750.0,
|
|
"step": 7270
|
|
},
|
|
{
|
|
"entropy": 5.37501106262207,
|
|
"epoch": 6.607905497501136,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001873313653074718,
|
|
"loss": 4.6959,
|
|
"mean_token_accuracy": 0.22417786568403245,
|
|
"num_tokens": 15056840.0,
|
|
"step": 7275
|
|
},
|
|
{
|
|
"entropy": 5.272376155853271,
|
|
"epoch": 6.612448886869605,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00018700597666235452,
|
|
"loss": 4.5578,
|
|
"mean_token_accuracy": 0.24203701615333556,
|
|
"num_tokens": 15067451.0,
|
|
"step": 7280
|
|
},
|
|
{
|
|
"entropy": 5.345329666137696,
|
|
"epoch": 6.616992276238074,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00018668080513374268,
|
|
"loss": 4.6039,
|
|
"mean_token_accuracy": 0.23896406590938568,
|
|
"num_tokens": 15077780.0,
|
|
"step": 7285
|
|
},
|
|
{
|
|
"entropy": 5.330894041061401,
|
|
"epoch": 6.621535665606542,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00018635585152396457,
|
|
"loss": 4.6794,
|
|
"mean_token_accuracy": 0.2262627437710762,
|
|
"num_tokens": 15089100.0,
|
|
"step": 7290
|
|
},
|
|
{
|
|
"entropy": 5.260492038726807,
|
|
"epoch": 6.626079054975011,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000186031116634811,
|
|
"loss": 4.5906,
|
|
"mean_token_accuracy": 0.2383026048541069,
|
|
"num_tokens": 15099420.0,
|
|
"step": 7295
|
|
},
|
|
{
|
|
"entropy": 5.375150871276856,
|
|
"epoch": 6.63062244434348,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00018570660126753291,
|
|
"loss": 4.6145,
|
|
"mean_token_accuracy": 0.23391713351011276,
|
|
"num_tokens": 15109884.0,
|
|
"step": 7300
|
|
},
|
|
{
|
|
"entropy": 5.317015504837036,
|
|
"epoch": 6.635165833711949,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001853823062228399,
|
|
"loss": 4.5482,
|
|
"mean_token_accuracy": 0.2473254159092903,
|
|
"num_tokens": 15119761.0,
|
|
"step": 7305
|
|
},
|
|
{
|
|
"entropy": 5.264550447463989,
|
|
"epoch": 6.639709223080418,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00018505823230089758,
|
|
"loss": 4.5364,
|
|
"mean_token_accuracy": 0.24828963875770568,
|
|
"num_tokens": 15129159.0,
|
|
"step": 7310
|
|
},
|
|
{
|
|
"entropy": 5.3173362731933596,
|
|
"epoch": 6.644252612448887,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00018473438030132633,
|
|
"loss": 4.5363,
|
|
"mean_token_accuracy": 0.2435192808508873,
|
|
"num_tokens": 15138855.0,
|
|
"step": 7315
|
|
},
|
|
{
|
|
"entropy": 5.326600170135498,
|
|
"epoch": 6.648796001817356,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001844107510231985,
|
|
"loss": 4.5854,
|
|
"mean_token_accuracy": 0.23622806072235109,
|
|
"num_tokens": 15149957.0,
|
|
"step": 7320
|
|
},
|
|
{
|
|
"entropy": 5.329568862915039,
|
|
"epoch": 6.653339391185825,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00018408734526503744,
|
|
"loss": 4.5953,
|
|
"mean_token_accuracy": 0.24421212822198868,
|
|
"num_tokens": 15159110.0,
|
|
"step": 7325
|
|
},
|
|
{
|
|
"entropy": 5.352579545974732,
|
|
"epoch": 6.657882780554294,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00018376416382481442,
|
|
"loss": 4.6675,
|
|
"mean_token_accuracy": 0.23189647793769835,
|
|
"num_tokens": 15169444.0,
|
|
"step": 7330
|
|
},
|
|
{
|
|
"entropy": 5.440320253372192,
|
|
"epoch": 6.662426169922762,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00018344120749994787,
|
|
"loss": 4.7889,
|
|
"mean_token_accuracy": 0.2206403613090515,
|
|
"num_tokens": 15180246.0,
|
|
"step": 7335
|
|
},
|
|
{
|
|
"entropy": 5.266606235504151,
|
|
"epoch": 6.666969559291231,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00018311847708730005,
|
|
"loss": 4.5515,
|
|
"mean_token_accuracy": 0.23758935034275055,
|
|
"num_tokens": 15191578.0,
|
|
"step": 7340
|
|
},
|
|
{
|
|
"entropy": 5.295640707015991,
|
|
"epoch": 6.6715129486597,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001827959733831765,
|
|
"loss": 4.5964,
|
|
"mean_token_accuracy": 0.23493965119123458,
|
|
"num_tokens": 15202380.0,
|
|
"step": 7345
|
|
},
|
|
{
|
|
"entropy": 5.261167335510254,
|
|
"epoch": 6.676056338028169,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00018247369718332274,
|
|
"loss": 4.5976,
|
|
"mean_token_accuracy": 0.23793933689594268,
|
|
"num_tokens": 15214073.0,
|
|
"step": 7350
|
|
},
|
|
{
|
|
"entropy": 5.285755252838134,
|
|
"epoch": 6.680599727396638,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00018215164928292354,
|
|
"loss": 4.5292,
|
|
"mean_token_accuracy": 0.242415352165699,
|
|
"num_tokens": 15223655.0,
|
|
"step": 7355
|
|
},
|
|
{
|
|
"entropy": 5.4032361030578615,
|
|
"epoch": 6.685143116765107,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00018182983047659984,
|
|
"loss": 4.7049,
|
|
"mean_token_accuracy": 0.22377368956804275,
|
|
"num_tokens": 15235567.0,
|
|
"step": 7360
|
|
},
|
|
{
|
|
"entropy": 5.334074211120606,
|
|
"epoch": 6.689686506133576,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00018150824155840778,
|
|
"loss": 4.5987,
|
|
"mean_token_accuracy": 0.23798300474882125,
|
|
"num_tokens": 15246349.0,
|
|
"step": 7365
|
|
},
|
|
{
|
|
"entropy": 5.3686279296875,
|
|
"epoch": 6.694229895502044,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001811868833218358,
|
|
"loss": 4.6513,
|
|
"mean_token_accuracy": 0.24255682826042174,
|
|
"num_tokens": 15256443.0,
|
|
"step": 7370
|
|
},
|
|
{
|
|
"entropy": 5.290392112731934,
|
|
"epoch": 6.698773284870513,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00018086575655980376,
|
|
"loss": 4.5829,
|
|
"mean_token_accuracy": 0.23714193403720857,
|
|
"num_tokens": 15267392.0,
|
|
"step": 7375
|
|
},
|
|
{
|
|
"entropy": 5.307732677459716,
|
|
"epoch": 6.703316674238982,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00018054486206465952,
|
|
"loss": 4.5484,
|
|
"mean_token_accuracy": 0.24576050490140916,
|
|
"num_tokens": 15277371.0,
|
|
"step": 7380
|
|
},
|
|
{
|
|
"entropy": 5.337274742126465,
|
|
"epoch": 6.707860063607451,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001802242006281788,
|
|
"loss": 4.5556,
|
|
"mean_token_accuracy": 0.23960385769605635,
|
|
"num_tokens": 15287490.0,
|
|
"step": 7385
|
|
},
|
|
{
|
|
"entropy": 5.2806800365447994,
|
|
"epoch": 6.71240345297592,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00017990377304156165,
|
|
"loss": 4.5653,
|
|
"mean_token_accuracy": 0.23982742726802825,
|
|
"num_tokens": 15298061.0,
|
|
"step": 7390
|
|
},
|
|
{
|
|
"entropy": 5.27966423034668,
|
|
"epoch": 6.716946842344389,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001795835800954312,
|
|
"loss": 4.5569,
|
|
"mean_token_accuracy": 0.24114361703395842,
|
|
"num_tokens": 15307862.0,
|
|
"step": 7395
|
|
},
|
|
{
|
|
"entropy": 5.2275467872619625,
|
|
"epoch": 6.721490231712858,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00017926362257983174,
|
|
"loss": 4.6225,
|
|
"mean_token_accuracy": 0.2407478615641594,
|
|
"num_tokens": 15317693.0,
|
|
"step": 7400
|
|
},
|
|
{
|
|
"entropy": 5.305719804763794,
|
|
"epoch": 6.726033621081327,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001789439012842269,
|
|
"loss": 4.6499,
|
|
"mean_token_accuracy": 0.2350459173321724,
|
|
"num_tokens": 15329000.0,
|
|
"step": 7405
|
|
},
|
|
{
|
|
"entropy": 5.347974824905395,
|
|
"epoch": 6.730577010449796,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00017862441699749687,
|
|
"loss": 4.6178,
|
|
"mean_token_accuracy": 0.24022860527038575,
|
|
"num_tokens": 15339967.0,
|
|
"step": 7410
|
|
},
|
|
{
|
|
"entropy": 5.363796854019165,
|
|
"epoch": 6.735120399818264,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001783051705079375,
|
|
"loss": 4.645,
|
|
"mean_token_accuracy": 0.23852888494729996,
|
|
"num_tokens": 15350478.0,
|
|
"step": 7415
|
|
},
|
|
{
|
|
"entropy": 5.368247699737549,
|
|
"epoch": 6.739663789186733,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00017798616260325767,
|
|
"loss": 4.5912,
|
|
"mean_token_accuracy": 0.23896664083004,
|
|
"num_tokens": 15360728.0,
|
|
"step": 7420
|
|
},
|
|
{
|
|
"entropy": 5.334335041046143,
|
|
"epoch": 6.744207178555202,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001776673940705778,
|
|
"loss": 4.6195,
|
|
"mean_token_accuracy": 0.23925431072711945,
|
|
"num_tokens": 15370689.0,
|
|
"step": 7425
|
|
},
|
|
{
|
|
"entropy": 5.288421583175659,
|
|
"epoch": 6.748750567923671,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001773488656964276,
|
|
"loss": 4.5276,
|
|
"mean_token_accuracy": 0.24424318224191666,
|
|
"num_tokens": 15380987.0,
|
|
"step": 7430
|
|
},
|
|
{
|
|
"entropy": 5.373984479904175,
|
|
"epoch": 6.75329395729214,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000177030578266744,
|
|
"loss": 4.6143,
|
|
"mean_token_accuracy": 0.2384542852640152,
|
|
"num_tokens": 15391989.0,
|
|
"step": 7435
|
|
},
|
|
{
|
|
"entropy": 5.339956569671631,
|
|
"epoch": 6.757837346660609,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017671253256686963,
|
|
"loss": 4.6389,
|
|
"mean_token_accuracy": 0.23211940824985505,
|
|
"num_tokens": 15402376.0,
|
|
"step": 7440
|
|
},
|
|
{
|
|
"entropy": 5.299764776229859,
|
|
"epoch": 6.762380736029078,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00017639472938155075,
|
|
"loss": 4.5811,
|
|
"mean_token_accuracy": 0.2337718576192856,
|
|
"num_tokens": 15413034.0,
|
|
"step": 7445
|
|
},
|
|
{
|
|
"entropy": 5.25767936706543,
|
|
"epoch": 6.766924125397547,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00017607716949493514,
|
|
"loss": 4.5374,
|
|
"mean_token_accuracy": 0.23684685230255126,
|
|
"num_tokens": 15423378.0,
|
|
"step": 7450
|
|
},
|
|
{
|
|
"entropy": 5.249557209014893,
|
|
"epoch": 6.771467514766016,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00017575985369057008,
|
|
"loss": 4.5367,
|
|
"mean_token_accuracy": 0.23672058135271073,
|
|
"num_tokens": 15434092.0,
|
|
"step": 7455
|
|
},
|
|
{
|
|
"entropy": 5.261110401153564,
|
|
"epoch": 6.776010904134484,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001754427827514008,
|
|
"loss": 4.5385,
|
|
"mean_token_accuracy": 0.23905250430107117,
|
|
"num_tokens": 15445178.0,
|
|
"step": 7460
|
|
},
|
|
{
|
|
"entropy": 5.341504621505737,
|
|
"epoch": 6.780554293502953,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00017512595745976855,
|
|
"loss": 4.6087,
|
|
"mean_token_accuracy": 0.23831457942724227,
|
|
"num_tokens": 15455456.0,
|
|
"step": 7465
|
|
},
|
|
{
|
|
"entropy": 5.282718753814697,
|
|
"epoch": 6.785097682871422,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00017480937859740793,
|
|
"loss": 4.568,
|
|
"mean_token_accuracy": 0.24525401443243028,
|
|
"num_tokens": 15465805.0,
|
|
"step": 7470
|
|
},
|
|
{
|
|
"entropy": 5.340976905822754,
|
|
"epoch": 6.789641072239891,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00017449304694544608,
|
|
"loss": 4.6221,
|
|
"mean_token_accuracy": 0.23625671118497849,
|
|
"num_tokens": 15476590.0,
|
|
"step": 7475
|
|
},
|
|
{
|
|
"entropy": 5.29100489616394,
|
|
"epoch": 6.79418446160836,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00017417696328439974,
|
|
"loss": 4.5701,
|
|
"mean_token_accuracy": 0.2382108449935913,
|
|
"num_tokens": 15486791.0,
|
|
"step": 7480
|
|
},
|
|
{
|
|
"entropy": 5.3179138660430905,
|
|
"epoch": 6.798727850976829,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00017386112839417402,
|
|
"loss": 4.5577,
|
|
"mean_token_accuracy": 0.24538885653018952,
|
|
"num_tokens": 15497116.0,
|
|
"step": 7485
|
|
},
|
|
{
|
|
"entropy": 5.328512859344483,
|
|
"epoch": 6.803271240345298,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00017354554305405994,
|
|
"loss": 4.5828,
|
|
"mean_token_accuracy": 0.24722437113523482,
|
|
"num_tokens": 15506731.0,
|
|
"step": 7490
|
|
},
|
|
{
|
|
"entropy": 5.290050458908081,
|
|
"epoch": 6.807814629713766,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001732302080427332,
|
|
"loss": 4.6148,
|
|
"mean_token_accuracy": 0.23462824672460555,
|
|
"num_tokens": 15517333.0,
|
|
"step": 7495
|
|
},
|
|
{
|
|
"entropy": 5.259932231903076,
|
|
"epoch": 6.812358019082235,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00017291512413825148,
|
|
"loss": 4.5799,
|
|
"mean_token_accuracy": 0.2396708607673645,
|
|
"num_tokens": 15526995.0,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"epoch": 6.812358019082235,
|
|
"eval_entropy": 5.1915477364562275,
|
|
"eval_loss": 5.65944766998291,
|
|
"eval_mean_token_accuracy": 0.17501534571827845,
|
|
"eval_num_tokens": 15526995.0,
|
|
"eval_runtime": 2.1908,
|
|
"eval_samples_per_second": 1567.893,
|
|
"eval_steps_per_second": 196.272,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"entropy": 5.309394359588623,
|
|
"epoch": 6.816901408450704,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00017260029211805297,
|
|
"loss": 4.5521,
|
|
"mean_token_accuracy": 0.23374093323946,
|
|
"num_tokens": 15536686.0,
|
|
"step": 7505
|
|
},
|
|
{
|
|
"entropy": 5.372293519973755,
|
|
"epoch": 6.821444797819173,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00017228571275895434,
|
|
"loss": 4.6876,
|
|
"mean_token_accuracy": 0.22446418255567552,
|
|
"num_tokens": 15547565.0,
|
|
"step": 7510
|
|
},
|
|
{
|
|
"entropy": 5.2536533832550045,
|
|
"epoch": 6.825988187187642,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00017197138683714902,
|
|
"loss": 4.5555,
|
|
"mean_token_accuracy": 0.23902540355920793,
|
|
"num_tokens": 15557962.0,
|
|
"step": 7515
|
|
},
|
|
{
|
|
"entropy": 5.296814584732056,
|
|
"epoch": 6.830531576556111,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001716573151282049,
|
|
"loss": 4.5872,
|
|
"mean_token_accuracy": 0.24163987785577773,
|
|
"num_tokens": 15567164.0,
|
|
"step": 7520
|
|
},
|
|
{
|
|
"entropy": 5.258332872390747,
|
|
"epoch": 6.83507496592458,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001713434984070627,
|
|
"loss": 4.5514,
|
|
"mean_token_accuracy": 0.23428046107292175,
|
|
"num_tokens": 15577917.0,
|
|
"step": 7525
|
|
},
|
|
{
|
|
"entropy": 5.3428874015808105,
|
|
"epoch": 6.839618355293049,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00017102993744803392,
|
|
"loss": 4.6456,
|
|
"mean_token_accuracy": 0.23314459770917892,
|
|
"num_tokens": 15589106.0,
|
|
"step": 7530
|
|
},
|
|
{
|
|
"entropy": 5.308885717391968,
|
|
"epoch": 6.844161744661518,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00017071663302479923,
|
|
"loss": 4.6584,
|
|
"mean_token_accuracy": 0.22871748358011246,
|
|
"num_tokens": 15600368.0,
|
|
"step": 7535
|
|
},
|
|
{
|
|
"entropy": 5.349568843841553,
|
|
"epoch": 6.848705134029986,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00017040358591040606,
|
|
"loss": 4.5726,
|
|
"mean_token_accuracy": 0.23840045183897018,
|
|
"num_tokens": 15610671.0,
|
|
"step": 7540
|
|
},
|
|
{
|
|
"entropy": 5.2473503112792965,
|
|
"epoch": 6.853248523398455,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000170090796877267,
|
|
"loss": 4.5259,
|
|
"mean_token_accuracy": 0.24578416347503662,
|
|
"num_tokens": 15620403.0,
|
|
"step": 7545
|
|
},
|
|
{
|
|
"entropy": 5.359851837158203,
|
|
"epoch": 6.857791912766924,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00016977826669715783,
|
|
"loss": 4.5888,
|
|
"mean_token_accuracy": 0.24295967817306519,
|
|
"num_tokens": 15631260.0,
|
|
"step": 7550
|
|
},
|
|
{
|
|
"entropy": 5.297138404846192,
|
|
"epoch": 6.862335302135393,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000169465996141216,
|
|
"loss": 4.5935,
|
|
"mean_token_accuracy": 0.23755543828010559,
|
|
"num_tokens": 15641604.0,
|
|
"step": 7555
|
|
},
|
|
{
|
|
"entropy": 5.329033613204956,
|
|
"epoch": 6.866878691503862,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001691539859799377,
|
|
"loss": 4.6028,
|
|
"mean_token_accuracy": 0.23715244382619857,
|
|
"num_tokens": 15652469.0,
|
|
"step": 7560
|
|
},
|
|
{
|
|
"entropy": 5.280193567276001,
|
|
"epoch": 6.871422080872331,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00016884223698317722,
|
|
"loss": 4.6056,
|
|
"mean_token_accuracy": 0.2331514075398445,
|
|
"num_tokens": 15662615.0,
|
|
"step": 7565
|
|
},
|
|
{
|
|
"entropy": 5.279680967330933,
|
|
"epoch": 6.8759654702408,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001685307499201441,
|
|
"loss": 4.5545,
|
|
"mean_token_accuracy": 0.23138108253479003,
|
|
"num_tokens": 15674868.0,
|
|
"step": 7570
|
|
},
|
|
{
|
|
"entropy": 5.328206348419189,
|
|
"epoch": 6.880508859609268,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001682195255594019,
|
|
"loss": 4.6446,
|
|
"mean_token_accuracy": 0.23994228094816208,
|
|
"num_tokens": 15686246.0,
|
|
"step": 7575
|
|
},
|
|
{
|
|
"entropy": 5.326714706420899,
|
|
"epoch": 6.885052248977737,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00016790856466886544,
|
|
"loss": 4.5688,
|
|
"mean_token_accuracy": 0.23696329593658447,
|
|
"num_tokens": 15697025.0,
|
|
"step": 7580
|
|
},
|
|
{
|
|
"entropy": 5.3247425079345705,
|
|
"epoch": 6.889595638346206,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016759786801580018,
|
|
"loss": 4.6359,
|
|
"mean_token_accuracy": 0.2321282371878624,
|
|
"num_tokens": 15706636.0,
|
|
"step": 7585
|
|
},
|
|
{
|
|
"entropy": 5.284221553802491,
|
|
"epoch": 6.894139027714675,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00016728743636681897,
|
|
"loss": 4.5731,
|
|
"mean_token_accuracy": 0.23938645273447037,
|
|
"num_tokens": 15717319.0,
|
|
"step": 7590
|
|
},
|
|
{
|
|
"entropy": 5.288644409179687,
|
|
"epoch": 6.898682417083144,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00016697727048788118,
|
|
"loss": 4.4982,
|
|
"mean_token_accuracy": 0.24991946667432785,
|
|
"num_tokens": 15726546.0,
|
|
"step": 7595
|
|
},
|
|
{
|
|
"entropy": 5.290557241439819,
|
|
"epoch": 6.903225806451613,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00016666737114429017,
|
|
"loss": 4.6168,
|
|
"mean_token_accuracy": 0.23205698877573014,
|
|
"num_tokens": 15737176.0,
|
|
"step": 7600
|
|
},
|
|
{
|
|
"entropy": 5.312217426300049,
|
|
"epoch": 6.907769195820082,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001663577391006919,
|
|
"loss": 4.596,
|
|
"mean_token_accuracy": 0.2422843798995018,
|
|
"num_tokens": 15747422.0,
|
|
"step": 7605
|
|
},
|
|
{
|
|
"entropy": 5.3302332878112795,
|
|
"epoch": 6.912312585188551,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00016604837512107263,
|
|
"loss": 4.6108,
|
|
"mean_token_accuracy": 0.22835036665201186,
|
|
"num_tokens": 15757591.0,
|
|
"step": 7610
|
|
},
|
|
{
|
|
"entropy": 5.310723924636841,
|
|
"epoch": 6.91685597455702,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00016573927996875714,
|
|
"loss": 4.6489,
|
|
"mean_token_accuracy": 0.23551349937915803,
|
|
"num_tokens": 15768375.0,
|
|
"step": 7615
|
|
},
|
|
{
|
|
"entropy": 5.305994176864624,
|
|
"epoch": 6.921399363925488,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00016543045440640697,
|
|
"loss": 4.5394,
|
|
"mean_token_accuracy": 0.24030639976263046,
|
|
"num_tokens": 15778935.0,
|
|
"step": 7620
|
|
},
|
|
{
|
|
"entropy": 5.311252927780151,
|
|
"epoch": 6.925942753293957,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00016512189919601865,
|
|
"loss": 4.5519,
|
|
"mean_token_accuracy": 0.24470688849687577,
|
|
"num_tokens": 15788955.0,
|
|
"step": 7625
|
|
},
|
|
{
|
|
"entropy": 5.352472162246704,
|
|
"epoch": 6.930486142662426,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001648136150989214,
|
|
"loss": 4.5449,
|
|
"mean_token_accuracy": 0.2497131869196892,
|
|
"num_tokens": 15798795.0,
|
|
"step": 7630
|
|
},
|
|
{
|
|
"entropy": 5.286505794525146,
|
|
"epoch": 6.935029532030895,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001645056028757756,
|
|
"loss": 4.6143,
|
|
"mean_token_accuracy": 0.23387709259986877,
|
|
"num_tokens": 15809251.0,
|
|
"step": 7635
|
|
},
|
|
{
|
|
"entropy": 5.2742249965667725,
|
|
"epoch": 6.939572921399364,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00016419786328657066,
|
|
"loss": 4.5275,
|
|
"mean_token_accuracy": 0.24301324635744095,
|
|
"num_tokens": 15819418.0,
|
|
"step": 7640
|
|
},
|
|
{
|
|
"entropy": 5.306789779663086,
|
|
"epoch": 6.944116310767833,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00016389039709062376,
|
|
"loss": 4.5827,
|
|
"mean_token_accuracy": 0.23372713327407837,
|
|
"num_tokens": 15829933.0,
|
|
"step": 7645
|
|
},
|
|
{
|
|
"entropy": 5.25427656173706,
|
|
"epoch": 6.948659700136302,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00016358320504657677,
|
|
"loss": 4.5953,
|
|
"mean_token_accuracy": 0.24130952656269072,
|
|
"num_tokens": 15841032.0,
|
|
"step": 7650
|
|
},
|
|
{
|
|
"entropy": 5.329875087738037,
|
|
"epoch": 6.95320308950477,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00016327628791239576,
|
|
"loss": 4.6397,
|
|
"mean_token_accuracy": 0.2313717171549797,
|
|
"num_tokens": 15851126.0,
|
|
"step": 7655
|
|
},
|
|
{
|
|
"entropy": 5.346979141235352,
|
|
"epoch": 6.957746478873239,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00016296964644536816,
|
|
"loss": 4.5991,
|
|
"mean_token_accuracy": 0.23252443373203277,
|
|
"num_tokens": 15861832.0,
|
|
"step": 7660
|
|
},
|
|
{
|
|
"entropy": 5.340782880783081,
|
|
"epoch": 6.962289868241708,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0001626632814021015,
|
|
"loss": 4.5942,
|
|
"mean_token_accuracy": 0.2359255388379097,
|
|
"num_tokens": 15872295.0,
|
|
"step": 7665
|
|
},
|
|
{
|
|
"entropy": 5.337462854385376,
|
|
"epoch": 6.966833257610177,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00016235719353852079,
|
|
"loss": 4.6021,
|
|
"mean_token_accuracy": 0.24071893393993377,
|
|
"num_tokens": 15882453.0,
|
|
"step": 7670
|
|
},
|
|
{
|
|
"entropy": 5.313927030563354,
|
|
"epoch": 6.971376646978646,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00016205138360986753,
|
|
"loss": 4.6211,
|
|
"mean_token_accuracy": 0.2359257698059082,
|
|
"num_tokens": 15891962.0,
|
|
"step": 7675
|
|
},
|
|
{
|
|
"entropy": 5.316891098022461,
|
|
"epoch": 6.975920036347115,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00016174585237069738,
|
|
"loss": 4.5317,
|
|
"mean_token_accuracy": 0.23896198570728303,
|
|
"num_tokens": 15902309.0,
|
|
"step": 7680
|
|
},
|
|
{
|
|
"entropy": 5.27788233757019,
|
|
"epoch": 6.980463425715584,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00016144060057487824,
|
|
"loss": 4.5503,
|
|
"mean_token_accuracy": 0.23794778883457185,
|
|
"num_tokens": 15912789.0,
|
|
"step": 7685
|
|
},
|
|
{
|
|
"entropy": 5.332029390335083,
|
|
"epoch": 6.985006815084053,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00016113562897558849,
|
|
"loss": 4.6302,
|
|
"mean_token_accuracy": 0.23401553481817244,
|
|
"num_tokens": 15922722.0,
|
|
"step": 7690
|
|
},
|
|
{
|
|
"entropy": 5.286727237701416,
|
|
"epoch": 6.989550204452522,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001608309383253154,
|
|
"loss": 4.5146,
|
|
"mean_token_accuracy": 0.24099305123090745,
|
|
"num_tokens": 15933069.0,
|
|
"step": 7695
|
|
},
|
|
{
|
|
"entropy": 5.35459041595459,
|
|
"epoch": 6.99409359382099,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001605265293758528,
|
|
"loss": 4.6024,
|
|
"mean_token_accuracy": 0.23342681676149368,
|
|
"num_tokens": 15942427.0,
|
|
"step": 7700
|
|
},
|
|
{
|
|
"entropy": 5.317014408111572,
|
|
"epoch": 6.998636983189459,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00016022240287829947,
|
|
"loss": 4.5854,
|
|
"mean_token_accuracy": 0.2365894705057144,
|
|
"num_tokens": 15952765.0,
|
|
"step": 7705
|
|
},
|
|
{
|
|
"entropy": 5.275751484764947,
|
|
"epoch": 7.002726033621081,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00015991855958305724,
|
|
"loss": 4.5613,
|
|
"mean_token_accuracy": 0.24785664512051475,
|
|
"num_tokens": 15963025.0,
|
|
"step": 7710
|
|
},
|
|
{
|
|
"entropy": 5.277569723129273,
|
|
"epoch": 7.00726942298955,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00015961500023982945,
|
|
"loss": 4.4953,
|
|
"mean_token_accuracy": 0.2477351412177086,
|
|
"num_tokens": 15973425.0,
|
|
"step": 7715
|
|
},
|
|
{
|
|
"entropy": 5.322409915924072,
|
|
"epoch": 7.011812812358019,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00015931172559761848,
|
|
"loss": 4.5102,
|
|
"mean_token_accuracy": 0.25534540265798567,
|
|
"num_tokens": 15984064.0,
|
|
"step": 7720
|
|
},
|
|
{
|
|
"entropy": 5.323396396636963,
|
|
"epoch": 7.016356201726488,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001590087364047243,
|
|
"loss": 4.5004,
|
|
"mean_token_accuracy": 0.2498973324894905,
|
|
"num_tokens": 15994512.0,
|
|
"step": 7725
|
|
},
|
|
{
|
|
"entropy": 5.280578517913819,
|
|
"epoch": 7.020899591094957,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00015870603340874267,
|
|
"loss": 4.4397,
|
|
"mean_token_accuracy": 0.24743602871894838,
|
|
"num_tokens": 16004140.0,
|
|
"step": 7730
|
|
},
|
|
{
|
|
"entropy": 5.2548261165618895,
|
|
"epoch": 7.025442980463426,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00015840361735656333,
|
|
"loss": 4.4291,
|
|
"mean_token_accuracy": 0.25202235132455825,
|
|
"num_tokens": 16013710.0,
|
|
"step": 7735
|
|
},
|
|
{
|
|
"entropy": 5.372058963775634,
|
|
"epoch": 7.029986369831895,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00015810148899436747,
|
|
"loss": 4.501,
|
|
"mean_token_accuracy": 0.25149954855442047,
|
|
"num_tokens": 16023846.0,
|
|
"step": 7740
|
|
},
|
|
{
|
|
"entropy": 5.294212436676025,
|
|
"epoch": 7.034529759200363,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00015779964906762706,
|
|
"loss": 4.4508,
|
|
"mean_token_accuracy": 0.24732972234487532,
|
|
"num_tokens": 16034327.0,
|
|
"step": 7745
|
|
},
|
|
{
|
|
"entropy": 5.277709674835205,
|
|
"epoch": 7.039073148568832,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00015749809832110197,
|
|
"loss": 4.5031,
|
|
"mean_token_accuracy": 0.25109738707542417,
|
|
"num_tokens": 16045055.0,
|
|
"step": 7750
|
|
},
|
|
{
|
|
"entropy": 5.239346027374268,
|
|
"epoch": 7.043616537937301,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001571968374988389,
|
|
"loss": 4.483,
|
|
"mean_token_accuracy": 0.24251957535743712,
|
|
"num_tokens": 16055198.0,
|
|
"step": 7755
|
|
},
|
|
{
|
|
"entropy": 5.246415424346924,
|
|
"epoch": 7.04815992730577,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00015689586734416876,
|
|
"loss": 4.4298,
|
|
"mean_token_accuracy": 0.25878233164548875,
|
|
"num_tokens": 16065030.0,
|
|
"step": 7760
|
|
},
|
|
{
|
|
"entropy": 5.33224310874939,
|
|
"epoch": 7.052703316674239,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00015659518859970545,
|
|
"loss": 4.5305,
|
|
"mean_token_accuracy": 0.25335928052663803,
|
|
"num_tokens": 16075974.0,
|
|
"step": 7765
|
|
},
|
|
{
|
|
"entropy": 5.321562623977661,
|
|
"epoch": 7.057246706042708,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001562948020073441,
|
|
"loss": 4.5362,
|
|
"mean_token_accuracy": 0.2489801526069641,
|
|
"num_tokens": 16086835.0,
|
|
"step": 7770
|
|
},
|
|
{
|
|
"entropy": 5.315760755538941,
|
|
"epoch": 7.061790095411177,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00015599470830825868,
|
|
"loss": 4.5154,
|
|
"mean_token_accuracy": 0.2440416246652603,
|
|
"num_tokens": 16098116.0,
|
|
"step": 7775
|
|
},
|
|
{
|
|
"entropy": 5.33055305480957,
|
|
"epoch": 7.066333484779646,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00015569490824290054,
|
|
"loss": 4.5286,
|
|
"mean_token_accuracy": 0.23909554481506348,
|
|
"num_tokens": 16109185.0,
|
|
"step": 7780
|
|
},
|
|
{
|
|
"entropy": 5.288166284561157,
|
|
"epoch": 7.070876874148114,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00015539540255099643,
|
|
"loss": 4.4852,
|
|
"mean_token_accuracy": 0.25098420977592467,
|
|
"num_tokens": 16120025.0,
|
|
"step": 7785
|
|
},
|
|
{
|
|
"entropy": 5.3047552585601805,
|
|
"epoch": 7.075420263516583,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00015509619197154705,
|
|
"loss": 4.4558,
|
|
"mean_token_accuracy": 0.25715659111738204,
|
|
"num_tokens": 16131145.0,
|
|
"step": 7790
|
|
},
|
|
{
|
|
"entropy": 5.363806581497192,
|
|
"epoch": 7.079963652885052,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00015479727724282476,
|
|
"loss": 4.5325,
|
|
"mean_token_accuracy": 0.24711658507585527,
|
|
"num_tokens": 16141063.0,
|
|
"step": 7795
|
|
},
|
|
{
|
|
"entropy": 5.311065626144409,
|
|
"epoch": 7.084507042253521,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00015449865910237184,
|
|
"loss": 4.4598,
|
|
"mean_token_accuracy": 0.2502642825245857,
|
|
"num_tokens": 16151059.0,
|
|
"step": 7800
|
|
},
|
|
{
|
|
"entropy": 5.270321464538574,
|
|
"epoch": 7.08905043162199,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00015420033828699882,
|
|
"loss": 4.4717,
|
|
"mean_token_accuracy": 0.25674335062503817,
|
|
"num_tokens": 16160679.0,
|
|
"step": 7805
|
|
},
|
|
{
|
|
"entropy": 5.249357080459594,
|
|
"epoch": 7.093593820990459,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001539023155327829,
|
|
"loss": 4.4831,
|
|
"mean_token_accuracy": 0.24401130974292756,
|
|
"num_tokens": 16170269.0,
|
|
"step": 7810
|
|
},
|
|
{
|
|
"entropy": 5.269272089004517,
|
|
"epoch": 7.098137210358928,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00015360459157506525,
|
|
"loss": 4.4798,
|
|
"mean_token_accuracy": 0.24897579550743104,
|
|
"num_tokens": 16180496.0,
|
|
"step": 7815
|
|
},
|
|
{
|
|
"entropy": 5.313865518569946,
|
|
"epoch": 7.102680599727397,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00015330716714845048,
|
|
"loss": 4.4885,
|
|
"mean_token_accuracy": 0.24036985784769058,
|
|
"num_tokens": 16191043.0,
|
|
"step": 7820
|
|
},
|
|
{
|
|
"entropy": 5.374632024765015,
|
|
"epoch": 7.107223989095866,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001530100429868035,
|
|
"loss": 4.6332,
|
|
"mean_token_accuracy": 0.23587190061807634,
|
|
"num_tokens": 16201551.0,
|
|
"step": 7825
|
|
},
|
|
{
|
|
"entropy": 5.3261895179748535,
|
|
"epoch": 7.111767378464334,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00015271321982324892,
|
|
"loss": 4.5293,
|
|
"mean_token_accuracy": 0.2380489394068718,
|
|
"num_tokens": 16212928.0,
|
|
"step": 7830
|
|
},
|
|
{
|
|
"entropy": 5.265306186676026,
|
|
"epoch": 7.116310767832803,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00015241669839016818,
|
|
"loss": 4.4898,
|
|
"mean_token_accuracy": 0.2489958018064499,
|
|
"num_tokens": 16223233.0,
|
|
"step": 7835
|
|
},
|
|
{
|
|
"entropy": 5.307391262054443,
|
|
"epoch": 7.120854157201272,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00015212047941919854,
|
|
"loss": 4.5661,
|
|
"mean_token_accuracy": 0.23868730813264846,
|
|
"num_tokens": 16234902.0,
|
|
"step": 7840
|
|
},
|
|
{
|
|
"entropy": 5.300672817230224,
|
|
"epoch": 7.125397546569741,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00015182456364123086,
|
|
"loss": 4.5225,
|
|
"mean_token_accuracy": 0.24378852397203446,
|
|
"num_tokens": 16244846.0,
|
|
"step": 7845
|
|
},
|
|
{
|
|
"entropy": 5.285271263122558,
|
|
"epoch": 7.12994093593821,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00015152895178640788,
|
|
"loss": 4.4793,
|
|
"mean_token_accuracy": 0.25083998292684556,
|
|
"num_tokens": 16255067.0,
|
|
"step": 7850
|
|
},
|
|
{
|
|
"entropy": 5.298731136322021,
|
|
"epoch": 7.134484325306679,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001512336445841224,
|
|
"loss": 4.5039,
|
|
"mean_token_accuracy": 0.2442450612783432,
|
|
"num_tokens": 16265482.0,
|
|
"step": 7855
|
|
},
|
|
{
|
|
"entropy": 5.363097190856934,
|
|
"epoch": 7.139027714675148,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00015093864276301573,
|
|
"loss": 4.5171,
|
|
"mean_token_accuracy": 0.24662193208932875,
|
|
"num_tokens": 16275319.0,
|
|
"step": 7860
|
|
},
|
|
{
|
|
"entropy": 5.262824201583863,
|
|
"epoch": 7.143571104043617,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00015064394705097546,
|
|
"loss": 4.5014,
|
|
"mean_token_accuracy": 0.24230348765850068,
|
|
"num_tokens": 16284889.0,
|
|
"step": 7865
|
|
},
|
|
{
|
|
"entropy": 5.328980255126953,
|
|
"epoch": 7.148114493412085,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00015034955817513397,
|
|
"loss": 4.5284,
|
|
"mean_token_accuracy": 0.24859661757946014,
|
|
"num_tokens": 16295259.0,
|
|
"step": 7870
|
|
},
|
|
{
|
|
"entropy": 5.242529010772705,
|
|
"epoch": 7.152657882780554,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001500554768618665,
|
|
"loss": 4.4762,
|
|
"mean_token_accuracy": 0.2482871189713478,
|
|
"num_tokens": 16305489.0,
|
|
"step": 7875
|
|
},
|
|
{
|
|
"entropy": 5.244515085220337,
|
|
"epoch": 7.157201272149023,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00014976170383678954,
|
|
"loss": 4.4558,
|
|
"mean_token_accuracy": 0.25950678884983064,
|
|
"num_tokens": 16315167.0,
|
|
"step": 7880
|
|
},
|
|
{
|
|
"entropy": 5.257901763916015,
|
|
"epoch": 7.161744661517492,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00014946823982475883,
|
|
"loss": 4.4502,
|
|
"mean_token_accuracy": 0.25237158238887786,
|
|
"num_tokens": 16325059.0,
|
|
"step": 7885
|
|
},
|
|
{
|
|
"entropy": 5.346208667755127,
|
|
"epoch": 7.166288050885961,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00014917508554986754,
|
|
"loss": 4.564,
|
|
"mean_token_accuracy": 0.2455112025141716,
|
|
"num_tokens": 16336647.0,
|
|
"step": 7890
|
|
},
|
|
{
|
|
"entropy": 5.301130723953247,
|
|
"epoch": 7.17083144025443,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00014888224173544476,
|
|
"loss": 4.4928,
|
|
"mean_token_accuracy": 0.23881710767745973,
|
|
"num_tokens": 16346146.0,
|
|
"step": 7895
|
|
},
|
|
{
|
|
"entropy": 5.328935432434082,
|
|
"epoch": 7.175374829622899,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00014858970910405368,
|
|
"loss": 4.5731,
|
|
"mean_token_accuracy": 0.24481898695230483,
|
|
"num_tokens": 16356124.0,
|
|
"step": 7900
|
|
},
|
|
{
|
|
"entropy": 5.286603164672852,
|
|
"epoch": 7.179918218991368,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00014829748837748906,
|
|
"loss": 4.531,
|
|
"mean_token_accuracy": 0.2408418983221054,
|
|
"num_tokens": 16367197.0,
|
|
"step": 7905
|
|
},
|
|
{
|
|
"entropy": 5.2644298553466795,
|
|
"epoch": 7.184461608359836,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001480055802767769,
|
|
"loss": 4.504,
|
|
"mean_token_accuracy": 0.23938197046518325,
|
|
"num_tokens": 16376757.0,
|
|
"step": 7910
|
|
},
|
|
{
|
|
"entropy": 5.25711030960083,
|
|
"epoch": 7.189004997728305,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00014771398552217114,
|
|
"loss": 4.4418,
|
|
"mean_token_accuracy": 0.2549315571784973,
|
|
"num_tokens": 16387454.0,
|
|
"step": 7915
|
|
},
|
|
{
|
|
"entropy": 5.320233488082886,
|
|
"epoch": 7.193548387096774,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00014742270483315319,
|
|
"loss": 4.5041,
|
|
"mean_token_accuracy": 0.2459973454475403,
|
|
"num_tokens": 16397852.0,
|
|
"step": 7920
|
|
},
|
|
{
|
|
"entropy": 5.2818056583404545,
|
|
"epoch": 7.198091776465243,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00014713173892842878,
|
|
"loss": 4.4653,
|
|
"mean_token_accuracy": 0.25257920324802396,
|
|
"num_tokens": 16407752.0,
|
|
"step": 7925
|
|
},
|
|
{
|
|
"entropy": 5.251960563659668,
|
|
"epoch": 7.202635165833712,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00014684108852592765,
|
|
"loss": 4.5038,
|
|
"mean_token_accuracy": 0.24475623518228531,
|
|
"num_tokens": 16418500.0,
|
|
"step": 7930
|
|
},
|
|
{
|
|
"entropy": 5.307971572875976,
|
|
"epoch": 7.207178555202181,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00014655075434280064,
|
|
"loss": 4.5292,
|
|
"mean_token_accuracy": 0.24503552317619323,
|
|
"num_tokens": 16428706.0,
|
|
"step": 7935
|
|
},
|
|
{
|
|
"entropy": 5.361135578155517,
|
|
"epoch": 7.21172194457065,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00014626073709541853,
|
|
"loss": 4.5087,
|
|
"mean_token_accuracy": 0.24916277825832367,
|
|
"num_tokens": 16438184.0,
|
|
"step": 7940
|
|
},
|
|
{
|
|
"entropy": 5.267688798904419,
|
|
"epoch": 7.216265333939119,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00014597103749937,
|
|
"loss": 4.4905,
|
|
"mean_token_accuracy": 0.24919096380472183,
|
|
"num_tokens": 16449340.0,
|
|
"step": 7945
|
|
},
|
|
{
|
|
"entropy": 5.303730297088623,
|
|
"epoch": 7.220808723307587,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00014568165626946012,
|
|
"loss": 4.4687,
|
|
"mean_token_accuracy": 0.2410537540912628,
|
|
"num_tokens": 16459604.0,
|
|
"step": 7950
|
|
},
|
|
{
|
|
"entropy": 5.357534408569336,
|
|
"epoch": 7.225352112676056,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00014539259411970823,
|
|
"loss": 4.5901,
|
|
"mean_token_accuracy": 0.2433488965034485,
|
|
"num_tokens": 16470150.0,
|
|
"step": 7955
|
|
},
|
|
{
|
|
"entropy": 5.335689640045166,
|
|
"epoch": 7.229895502044525,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001451038517633465,
|
|
"loss": 4.5226,
|
|
"mean_token_accuracy": 0.25400001406669614,
|
|
"num_tokens": 16480797.0,
|
|
"step": 7960
|
|
},
|
|
{
|
|
"entropy": 5.3175952434539795,
|
|
"epoch": 7.234438891412994,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00014481542991281782,
|
|
"loss": 4.5789,
|
|
"mean_token_accuracy": 0.24218350499868393,
|
|
"num_tokens": 16492629.0,
|
|
"step": 7965
|
|
},
|
|
{
|
|
"entropy": 5.299832677841186,
|
|
"epoch": 7.238982280781463,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00014452732927977465,
|
|
"loss": 4.4406,
|
|
"mean_token_accuracy": 0.2448546275496483,
|
|
"num_tokens": 16503225.0,
|
|
"step": 7970
|
|
},
|
|
{
|
|
"entropy": 5.31223955154419,
|
|
"epoch": 7.243525670149932,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001442395505750766,
|
|
"loss": 4.4423,
|
|
"mean_token_accuracy": 0.25841270536184313,
|
|
"num_tokens": 16512446.0,
|
|
"step": 7975
|
|
},
|
|
{
|
|
"entropy": 5.256321907043457,
|
|
"epoch": 7.248069059518401,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00014395209450878899,
|
|
"loss": 4.4817,
|
|
"mean_token_accuracy": 0.24859133660793303,
|
|
"num_tokens": 16522944.0,
|
|
"step": 7980
|
|
},
|
|
{
|
|
"entropy": 5.2756775379180905,
|
|
"epoch": 7.25261244888687,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000143664961790181,
|
|
"loss": 4.4857,
|
|
"mean_token_accuracy": 0.24977014809846879,
|
|
"num_tokens": 16533100.0,
|
|
"step": 7985
|
|
},
|
|
{
|
|
"entropy": 5.315029764175415,
|
|
"epoch": 7.257155838255338,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001433781531277243,
|
|
"loss": 4.5546,
|
|
"mean_token_accuracy": 0.2392375111579895,
|
|
"num_tokens": 16542893.0,
|
|
"step": 7990
|
|
},
|
|
{
|
|
"entropy": 5.3025744438171385,
|
|
"epoch": 7.261699227623807,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001430916692290905,
|
|
"loss": 4.5043,
|
|
"mean_token_accuracy": 0.2463330090045929,
|
|
"num_tokens": 16553209.0,
|
|
"step": 7995
|
|
},
|
|
{
|
|
"entropy": 5.362533140182495,
|
|
"epoch": 7.266242616992276,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00014280551080115032,
|
|
"loss": 4.5198,
|
|
"mean_token_accuracy": 0.24981896132230758,
|
|
"num_tokens": 16564574.0,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"epoch": 7.266242616992276,
|
|
"eval_entropy": 5.168546892875849,
|
|
"eval_loss": 5.675045013427734,
|
|
"eval_mean_token_accuracy": 0.1746888506204583,
|
|
"eval_num_tokens": 16564574.0,
|
|
"eval_runtime": 1.9991,
|
|
"eval_samples_per_second": 1718.26,
|
|
"eval_steps_per_second": 215.095,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"entropy": 5.306761884689331,
|
|
"epoch": 7.270786006360745,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00014251967854997106,
|
|
"loss": 4.5234,
|
|
"mean_token_accuracy": 0.2451857089996338,
|
|
"num_tokens": 16574517.0,
|
|
"step": 8005
|
|
},
|
|
{
|
|
"entropy": 5.333418941497802,
|
|
"epoch": 7.275329395729214,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00014223417318081572,
|
|
"loss": 4.5332,
|
|
"mean_token_accuracy": 0.24349355548620225,
|
|
"num_tokens": 16585311.0,
|
|
"step": 8010
|
|
},
|
|
{
|
|
"entropy": 5.251169872283936,
|
|
"epoch": 7.279872785097683,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00014194899539814,
|
|
"loss": 4.5076,
|
|
"mean_token_accuracy": 0.24374237805604934,
|
|
"num_tokens": 16596055.0,
|
|
"step": 8015
|
|
},
|
|
{
|
|
"entropy": 5.210993099212646,
|
|
"epoch": 7.284416174466152,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00014166414590559197,
|
|
"loss": 4.4593,
|
|
"mean_token_accuracy": 0.2507786348462105,
|
|
"num_tokens": 16606076.0,
|
|
"step": 8020
|
|
},
|
|
{
|
|
"entropy": 5.323359823226928,
|
|
"epoch": 7.288959563834621,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00014137962540600944,
|
|
"loss": 4.5238,
|
|
"mean_token_accuracy": 0.2413215681910515,
|
|
"num_tokens": 16616740.0,
|
|
"step": 8025
|
|
},
|
|
{
|
|
"entropy": 5.284033966064453,
|
|
"epoch": 7.29350295320309,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00014109543460141837,
|
|
"loss": 4.4628,
|
|
"mean_token_accuracy": 0.2528661251068115,
|
|
"num_tokens": 16627412.0,
|
|
"step": 8030
|
|
},
|
|
{
|
|
"entropy": 5.273031616210938,
|
|
"epoch": 7.298046342571558,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001408115741930313,
|
|
"loss": 4.4822,
|
|
"mean_token_accuracy": 0.2504570946097374,
|
|
"num_tokens": 16637032.0,
|
|
"step": 8035
|
|
},
|
|
{
|
|
"entropy": 5.246134805679321,
|
|
"epoch": 7.302589731940027,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00014052804488124575,
|
|
"loss": 4.4746,
|
|
"mean_token_accuracy": 0.2452763557434082,
|
|
"num_tokens": 16646596.0,
|
|
"step": 8040
|
|
},
|
|
{
|
|
"entropy": 5.280140113830567,
|
|
"epoch": 7.307133121308496,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014024484736564196,
|
|
"loss": 4.4241,
|
|
"mean_token_accuracy": 0.2541095092892647,
|
|
"num_tokens": 16656404.0,
|
|
"step": 8045
|
|
},
|
|
{
|
|
"entropy": 5.280656576156616,
|
|
"epoch": 7.311676510676965,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001399619823449817,
|
|
"loss": 4.4992,
|
|
"mean_token_accuracy": 0.246099790930748,
|
|
"num_tokens": 16666372.0,
|
|
"step": 8050
|
|
},
|
|
{
|
|
"entropy": 5.266527318954468,
|
|
"epoch": 7.316219900045434,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00013967945051720614,
|
|
"loss": 4.493,
|
|
"mean_token_accuracy": 0.24992201030254363,
|
|
"num_tokens": 16677150.0,
|
|
"step": 8055
|
|
},
|
|
{
|
|
"entropy": 5.278942537307739,
|
|
"epoch": 7.320763289413903,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00013939725257943475,
|
|
"loss": 4.5542,
|
|
"mean_token_accuracy": 0.24392787516117095,
|
|
"num_tokens": 16688712.0,
|
|
"step": 8060
|
|
},
|
|
{
|
|
"entropy": 5.293560600280761,
|
|
"epoch": 7.325306678782372,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00013911538922796275,
|
|
"loss": 4.5073,
|
|
"mean_token_accuracy": 0.24121430963277818,
|
|
"num_tokens": 16699214.0,
|
|
"step": 8065
|
|
},
|
|
{
|
|
"entropy": 5.271330165863037,
|
|
"epoch": 7.32985006815084,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00013883386115825984,
|
|
"loss": 4.4995,
|
|
"mean_token_accuracy": 0.24839711040258408,
|
|
"num_tokens": 16709294.0,
|
|
"step": 8070
|
|
},
|
|
{
|
|
"entropy": 5.270096254348755,
|
|
"epoch": 7.334393457519309,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00013855266906496867,
|
|
"loss": 4.4586,
|
|
"mean_token_accuracy": 0.2523455172777176,
|
|
"num_tokens": 16718761.0,
|
|
"step": 8075
|
|
},
|
|
{
|
|
"entropy": 5.317650699615479,
|
|
"epoch": 7.338936846887778,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00013827181364190288,
|
|
"loss": 4.5733,
|
|
"mean_token_accuracy": 0.24002302438020706,
|
|
"num_tokens": 16729075.0,
|
|
"step": 8080
|
|
},
|
|
{
|
|
"entropy": 5.283512353897095,
|
|
"epoch": 7.343480236256247,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00013799129558204505,
|
|
"loss": 4.4501,
|
|
"mean_token_accuracy": 0.25039338022470475,
|
|
"num_tokens": 16739293.0,
|
|
"step": 8085
|
|
},
|
|
{
|
|
"entropy": 5.287774276733399,
|
|
"epoch": 7.348023625624716,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00013771111557754577,
|
|
"loss": 4.5018,
|
|
"mean_token_accuracy": 0.2502776235342026,
|
|
"num_tokens": 16749214.0,
|
|
"step": 8090
|
|
},
|
|
{
|
|
"entropy": 5.246589040756225,
|
|
"epoch": 7.352567014993185,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00013743127431972124,
|
|
"loss": 4.4639,
|
|
"mean_token_accuracy": 0.24920261204242705,
|
|
"num_tokens": 16758715.0,
|
|
"step": 8095
|
|
},
|
|
{
|
|
"entropy": 5.286833667755127,
|
|
"epoch": 7.357110404361654,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001371517724990522,
|
|
"loss": 4.5455,
|
|
"mean_token_accuracy": 0.24820690006017684,
|
|
"num_tokens": 16768965.0,
|
|
"step": 8100
|
|
},
|
|
{
|
|
"entropy": 5.285393095016479,
|
|
"epoch": 7.361653793730123,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0001368726108051813,
|
|
"loss": 4.4037,
|
|
"mean_token_accuracy": 0.2578710734844208,
|
|
"num_tokens": 16778368.0,
|
|
"step": 8105
|
|
},
|
|
{
|
|
"entropy": 5.343227291107178,
|
|
"epoch": 7.366197183098592,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00013659378992691244,
|
|
"loss": 4.5304,
|
|
"mean_token_accuracy": 0.2449859008193016,
|
|
"num_tokens": 16789396.0,
|
|
"step": 8110
|
|
},
|
|
{
|
|
"entropy": 5.323299074172974,
|
|
"epoch": 7.37074057246706,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00013631531055220835,
|
|
"loss": 4.5317,
|
|
"mean_token_accuracy": 0.24524221420288086,
|
|
"num_tokens": 16800063.0,
|
|
"step": 8115
|
|
},
|
|
{
|
|
"entropy": 5.26822338104248,
|
|
"epoch": 7.375283961835529,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00013603717336818934,
|
|
"loss": 4.4835,
|
|
"mean_token_accuracy": 0.24528562128543854,
|
|
"num_tokens": 16810344.0,
|
|
"step": 8120
|
|
},
|
|
{
|
|
"entropy": 5.249145412445069,
|
|
"epoch": 7.379827351203998,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00013575937906113095,
|
|
"loss": 4.4482,
|
|
"mean_token_accuracy": 0.251977775990963,
|
|
"num_tokens": 16820233.0,
|
|
"step": 8125
|
|
},
|
|
{
|
|
"entropy": 5.287741231918335,
|
|
"epoch": 7.384370740572467,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0001354819283164634,
|
|
"loss": 4.5385,
|
|
"mean_token_accuracy": 0.24505277276039122,
|
|
"num_tokens": 16829761.0,
|
|
"step": 8130
|
|
},
|
|
{
|
|
"entropy": 5.333208751678467,
|
|
"epoch": 7.388914129940936,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001352048218187686,
|
|
"loss": 4.603,
|
|
"mean_token_accuracy": 0.2414749339222908,
|
|
"num_tokens": 16841068.0,
|
|
"step": 8135
|
|
},
|
|
{
|
|
"entropy": 5.296948099136353,
|
|
"epoch": 7.393457519309405,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00013492806025177935,
|
|
"loss": 4.4555,
|
|
"mean_token_accuracy": 0.25061228275299074,
|
|
"num_tokens": 16850733.0,
|
|
"step": 8140
|
|
},
|
|
{
|
|
"entropy": 5.2957500457763675,
|
|
"epoch": 7.398000908677874,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00013465164429837716,
|
|
"loss": 4.4954,
|
|
"mean_token_accuracy": 0.24421378523111342,
|
|
"num_tokens": 16860416.0,
|
|
"step": 8145
|
|
},
|
|
{
|
|
"entropy": 5.335061740875244,
|
|
"epoch": 7.402544298046343,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001343755746405912,
|
|
"loss": 4.5168,
|
|
"mean_token_accuracy": 0.25209966003894807,
|
|
"num_tokens": 16870662.0,
|
|
"step": 8150
|
|
},
|
|
{
|
|
"entropy": 5.2810962200164795,
|
|
"epoch": 7.407087687414811,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00013409985195959574,
|
|
"loss": 4.5683,
|
|
"mean_token_accuracy": 0.23571641743183136,
|
|
"num_tokens": 16881758.0,
|
|
"step": 8155
|
|
},
|
|
{
|
|
"entropy": 5.286536073684692,
|
|
"epoch": 7.41163107678328,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00013382447693570908,
|
|
"loss": 4.5354,
|
|
"mean_token_accuracy": 0.23793915212154387,
|
|
"num_tokens": 16892399.0,
|
|
"step": 8160
|
|
},
|
|
{
|
|
"entropy": 5.293103790283203,
|
|
"epoch": 7.416174466151749,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00013354945024839166,
|
|
"loss": 4.5612,
|
|
"mean_token_accuracy": 0.24520416110754012,
|
|
"num_tokens": 16903696.0,
|
|
"step": 8165
|
|
},
|
|
{
|
|
"entropy": 5.317911767959595,
|
|
"epoch": 7.420717855520218,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00013327477257624466,
|
|
"loss": 4.5103,
|
|
"mean_token_accuracy": 0.24359079748392104,
|
|
"num_tokens": 16914326.0,
|
|
"step": 8170
|
|
},
|
|
{
|
|
"entropy": 5.264439344406128,
|
|
"epoch": 7.425261244888687,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001330004445970078,
|
|
"loss": 4.4376,
|
|
"mean_token_accuracy": 0.25557203143835067,
|
|
"num_tokens": 16924533.0,
|
|
"step": 8175
|
|
},
|
|
{
|
|
"entropy": 5.290355443954468,
|
|
"epoch": 7.429804634257156,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00013272646698755816,
|
|
"loss": 4.4587,
|
|
"mean_token_accuracy": 0.2522106409072876,
|
|
"num_tokens": 16934157.0,
|
|
"step": 8180
|
|
},
|
|
{
|
|
"entropy": 5.211922597885132,
|
|
"epoch": 7.434348023625625,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00013245284042390814,
|
|
"loss": 4.42,
|
|
"mean_token_accuracy": 0.25636973083019254,
|
|
"num_tokens": 16944148.0,
|
|
"step": 8185
|
|
},
|
|
{
|
|
"entropy": 5.3079187870025635,
|
|
"epoch": 7.438891412994094,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00013217956558120436,
|
|
"loss": 4.5729,
|
|
"mean_token_accuracy": 0.23830318152904512,
|
|
"num_tokens": 16956388.0,
|
|
"step": 8190
|
|
},
|
|
{
|
|
"entropy": 5.286941576004028,
|
|
"epoch": 7.443434802362562,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.000131906643133725,
|
|
"loss": 4.5401,
|
|
"mean_token_accuracy": 0.24529509544372557,
|
|
"num_tokens": 16965216.0,
|
|
"step": 8195
|
|
},
|
|
{
|
|
"entropy": 5.283944129943848,
|
|
"epoch": 7.447978191731031,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00013163407375487926,
|
|
"loss": 4.4937,
|
|
"mean_token_accuracy": 0.248878875374794,
|
|
"num_tokens": 16976250.0,
|
|
"step": 8200
|
|
},
|
|
{
|
|
"entropy": 5.2779998779296875,
|
|
"epoch": 7.4525215810995,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00013136185811720496,
|
|
"loss": 4.5366,
|
|
"mean_token_accuracy": 0.24047518521547318,
|
|
"num_tokens": 16987375.0,
|
|
"step": 8205
|
|
},
|
|
{
|
|
"entropy": 5.235025644302368,
|
|
"epoch": 7.457064970467969,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001310899968923671,
|
|
"loss": 4.5288,
|
|
"mean_token_accuracy": 0.24733151942491532,
|
|
"num_tokens": 16997526.0,
|
|
"step": 8210
|
|
},
|
|
{
|
|
"entropy": 5.287834310531617,
|
|
"epoch": 7.461608359836438,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001308184907511562,
|
|
"loss": 4.5421,
|
|
"mean_token_accuracy": 0.24582349359989167,
|
|
"num_tokens": 17008342.0,
|
|
"step": 8215
|
|
},
|
|
{
|
|
"entropy": 5.28959641456604,
|
|
"epoch": 7.466151749204907,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00013054734036348675,
|
|
"loss": 4.5317,
|
|
"mean_token_accuracy": 0.2459393098950386,
|
|
"num_tokens": 17019555.0,
|
|
"step": 8220
|
|
},
|
|
{
|
|
"entropy": 5.296174764633179,
|
|
"epoch": 7.470695138573376,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00013027654639839536,
|
|
"loss": 4.5419,
|
|
"mean_token_accuracy": 0.2462657779455185,
|
|
"num_tokens": 17030686.0,
|
|
"step": 8225
|
|
},
|
|
{
|
|
"entropy": 5.3364753246307375,
|
|
"epoch": 7.475238527941845,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00013000610952403917,
|
|
"loss": 4.4967,
|
|
"mean_token_accuracy": 0.2534152403473854,
|
|
"num_tokens": 17041089.0,
|
|
"step": 8230
|
|
},
|
|
{
|
|
"entropy": 5.242576932907104,
|
|
"epoch": 7.479781917310314,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00012973603040769427,
|
|
"loss": 4.4501,
|
|
"mean_token_accuracy": 0.2467125967144966,
|
|
"num_tokens": 17050194.0,
|
|
"step": 8235
|
|
},
|
|
{
|
|
"entropy": 5.280106449127198,
|
|
"epoch": 7.484325306678782,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00012946630971575414,
|
|
"loss": 4.4967,
|
|
"mean_token_accuracy": 0.23874559849500657,
|
|
"num_tokens": 17060596.0,
|
|
"step": 8240
|
|
},
|
|
{
|
|
"entropy": 5.2974554061889645,
|
|
"epoch": 7.488868696047251,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00012919694811372768,
|
|
"loss": 4.5464,
|
|
"mean_token_accuracy": 0.24256813526153564,
|
|
"num_tokens": 17069763.0,
|
|
"step": 8245
|
|
},
|
|
{
|
|
"entropy": 5.234449005126953,
|
|
"epoch": 7.49341208541572,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00012892794626623793,
|
|
"loss": 4.4688,
|
|
"mean_token_accuracy": 0.24908677488565445,
|
|
"num_tokens": 17080031.0,
|
|
"step": 8250
|
|
},
|
|
{
|
|
"entropy": 5.271715450286865,
|
|
"epoch": 7.497955474784189,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001286593048370201,
|
|
"loss": 4.5102,
|
|
"mean_token_accuracy": 0.24316747039556502,
|
|
"num_tokens": 17091339.0,
|
|
"step": 8255
|
|
},
|
|
{
|
|
"entropy": 5.241258764266968,
|
|
"epoch": 7.502498864152658,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00012839102448892033,
|
|
"loss": 4.4626,
|
|
"mean_token_accuracy": 0.25320319533348085,
|
|
"num_tokens": 17102264.0,
|
|
"step": 8260
|
|
},
|
|
{
|
|
"entropy": 5.272535181045532,
|
|
"epoch": 7.507042253521127,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00012812310588389367,
|
|
"loss": 4.5103,
|
|
"mean_token_accuracy": 0.2443411111831665,
|
|
"num_tokens": 17112493.0,
|
|
"step": 8265
|
|
},
|
|
{
|
|
"entropy": 5.2961994171142575,
|
|
"epoch": 7.511585642889596,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00012785554968300266,
|
|
"loss": 4.4629,
|
|
"mean_token_accuracy": 0.2548535794019699,
|
|
"num_tokens": 17122037.0,
|
|
"step": 8270
|
|
},
|
|
{
|
|
"entropy": 5.286315822601319,
|
|
"epoch": 7.516129032258064,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00012758835654641548,
|
|
"loss": 4.5065,
|
|
"mean_token_accuracy": 0.25015697032213213,
|
|
"num_tokens": 17133403.0,
|
|
"step": 8275
|
|
},
|
|
{
|
|
"entropy": 5.224550104141235,
|
|
"epoch": 7.520672421626533,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000127321527133405,
|
|
"loss": 4.4571,
|
|
"mean_token_accuracy": 0.2531124919652939,
|
|
"num_tokens": 17143619.0,
|
|
"step": 8280
|
|
},
|
|
{
|
|
"entropy": 5.315179824829102,
|
|
"epoch": 7.525215810995002,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00012705506210234583,
|
|
"loss": 4.5535,
|
|
"mean_token_accuracy": 0.24149513244628906,
|
|
"num_tokens": 17154811.0,
|
|
"step": 8285
|
|
},
|
|
{
|
|
"entropy": 5.256762599945068,
|
|
"epoch": 7.529759200363471,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00012678896211071422,
|
|
"loss": 4.4704,
|
|
"mean_token_accuracy": 0.23869513869285583,
|
|
"num_tokens": 17166039.0,
|
|
"step": 8290
|
|
},
|
|
{
|
|
"entropy": 5.292482042312622,
|
|
"epoch": 7.53430258973194,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001265232278150854,
|
|
"loss": 4.4982,
|
|
"mean_token_accuracy": 0.24482963383197784,
|
|
"num_tokens": 17176555.0,
|
|
"step": 8295
|
|
},
|
|
{
|
|
"entropy": 5.266217565536499,
|
|
"epoch": 7.538845979100409,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001262578598711323,
|
|
"loss": 4.5386,
|
|
"mean_token_accuracy": 0.23710803985595702,
|
|
"num_tokens": 17188220.0,
|
|
"step": 8300
|
|
},
|
|
{
|
|
"entropy": 5.29193377494812,
|
|
"epoch": 7.543389368468878,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001259928589336239,
|
|
"loss": 4.5048,
|
|
"mean_token_accuracy": 0.2509712904691696,
|
|
"num_tokens": 17199528.0,
|
|
"step": 8305
|
|
},
|
|
{
|
|
"entropy": 5.286821174621582,
|
|
"epoch": 7.547932757837347,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00012572822565642378,
|
|
"loss": 4.4665,
|
|
"mean_token_accuracy": 0.2459220916032791,
|
|
"num_tokens": 17210030.0,
|
|
"step": 8310
|
|
},
|
|
{
|
|
"entropy": 5.296505451202393,
|
|
"epoch": 7.552476147205816,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00012546396069248813,
|
|
"loss": 4.5565,
|
|
"mean_token_accuracy": 0.24261924922466277,
|
|
"num_tokens": 17220231.0,
|
|
"step": 8315
|
|
},
|
|
{
|
|
"entropy": 5.26624641418457,
|
|
"epoch": 7.557019536574284,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00012520006469386458,
|
|
"loss": 4.5027,
|
|
"mean_token_accuracy": 0.24963923096656798,
|
|
"num_tokens": 17229966.0,
|
|
"step": 8320
|
|
},
|
|
{
|
|
"entropy": 5.309925174713134,
|
|
"epoch": 7.561562925942753,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00012493653831169014,
|
|
"loss": 4.5308,
|
|
"mean_token_accuracy": 0.24624732732772828,
|
|
"num_tokens": 17240654.0,
|
|
"step": 8325
|
|
},
|
|
{
|
|
"entropy": 5.226900482177735,
|
|
"epoch": 7.566106315311222,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00012467338219619015,
|
|
"loss": 4.4373,
|
|
"mean_token_accuracy": 0.2579703599214554,
|
|
"num_tokens": 17250653.0,
|
|
"step": 8330
|
|
},
|
|
{
|
|
"entropy": 5.320538282394409,
|
|
"epoch": 7.570649704679691,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.000124410596996676,
|
|
"loss": 4.5401,
|
|
"mean_token_accuracy": 0.252753846347332,
|
|
"num_tokens": 17260223.0,
|
|
"step": 8335
|
|
},
|
|
{
|
|
"entropy": 5.270241785049438,
|
|
"epoch": 7.57519309404816,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001241481833615442,
|
|
"loss": 4.4877,
|
|
"mean_token_accuracy": 0.2437804326415062,
|
|
"num_tokens": 17269998.0,
|
|
"step": 8340
|
|
},
|
|
{
|
|
"entropy": 5.258489799499512,
|
|
"epoch": 7.579736483416629,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00012388614193827407,
|
|
"loss": 4.4595,
|
|
"mean_token_accuracy": 0.25441806614398954,
|
|
"num_tokens": 17279824.0,
|
|
"step": 8345
|
|
},
|
|
{
|
|
"entropy": 5.249262714385987,
|
|
"epoch": 7.584279872785098,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000123624473373427,
|
|
"loss": 4.4736,
|
|
"mean_token_accuracy": 0.24940361380577086,
|
|
"num_tokens": 17290242.0,
|
|
"step": 8350
|
|
},
|
|
{
|
|
"entropy": 5.227822875976562,
|
|
"epoch": 7.588823262153566,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00012336317831264411,
|
|
"loss": 4.4544,
|
|
"mean_token_accuracy": 0.2550961419939995,
|
|
"num_tokens": 17299515.0,
|
|
"step": 8355
|
|
},
|
|
{
|
|
"entropy": 5.250787448883057,
|
|
"epoch": 7.593366651522036,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00012310225740064498,
|
|
"loss": 4.4375,
|
|
"mean_token_accuracy": 0.252620929479599,
|
|
"num_tokens": 17308627.0,
|
|
"step": 8360
|
|
},
|
|
{
|
|
"entropy": 5.21569037437439,
|
|
"epoch": 7.597910040890504,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001228417112812259,
|
|
"loss": 4.4167,
|
|
"mean_token_accuracy": 0.24488166719675064,
|
|
"num_tokens": 17319465.0,
|
|
"step": 8365
|
|
},
|
|
{
|
|
"entropy": 5.223727560043335,
|
|
"epoch": 7.602453430258973,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00012258154059725877,
|
|
"loss": 4.4676,
|
|
"mean_token_accuracy": 0.2488279238343239,
|
|
"num_tokens": 17329202.0,
|
|
"step": 8370
|
|
},
|
|
{
|
|
"entropy": 5.217889499664307,
|
|
"epoch": 7.606996819627442,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0001223217459906886,
|
|
"loss": 4.4235,
|
|
"mean_token_accuracy": 0.2559898689389229,
|
|
"num_tokens": 17338956.0,
|
|
"step": 8375
|
|
},
|
|
{
|
|
"entropy": 5.332046127319336,
|
|
"epoch": 7.611540208995911,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00012206232810253297,
|
|
"loss": 4.5876,
|
|
"mean_token_accuracy": 0.24184440225362777,
|
|
"num_tokens": 17349380.0,
|
|
"step": 8380
|
|
},
|
|
{
|
|
"entropy": 5.312648057937622,
|
|
"epoch": 7.61608359836438,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00012180328757287967,
|
|
"loss": 4.6197,
|
|
"mean_token_accuracy": 0.23640539944171907,
|
|
"num_tokens": 17359770.0,
|
|
"step": 8385
|
|
},
|
|
{
|
|
"entropy": 5.251143312454223,
|
|
"epoch": 7.620626987732849,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00012154462504088545,
|
|
"loss": 4.4707,
|
|
"mean_token_accuracy": 0.2501197010278702,
|
|
"num_tokens": 17369209.0,
|
|
"step": 8390
|
|
},
|
|
{
|
|
"entropy": 5.244078350067139,
|
|
"epoch": 7.625170377101318,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001212863411447744,
|
|
"loss": 4.4955,
|
|
"mean_token_accuracy": 0.24469618499279022,
|
|
"num_tokens": 17379221.0,
|
|
"step": 8395
|
|
},
|
|
{
|
|
"entropy": 5.3013999462127686,
|
|
"epoch": 7.629713766469786,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00012102843652183628,
|
|
"loss": 4.4281,
|
|
"mean_token_accuracy": 0.2524171069264412,
|
|
"num_tokens": 17389164.0,
|
|
"step": 8400
|
|
},
|
|
{
|
|
"entropy": 5.295888805389405,
|
|
"epoch": 7.634257155838255,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00012077091180842523,
|
|
"loss": 4.5302,
|
|
"mean_token_accuracy": 0.24815105497837067,
|
|
"num_tokens": 17399392.0,
|
|
"step": 8405
|
|
},
|
|
{
|
|
"entropy": 5.313620138168335,
|
|
"epoch": 7.638800545206724,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00012051376763995784,
|
|
"loss": 4.5064,
|
|
"mean_token_accuracy": 0.24525736272335052,
|
|
"num_tokens": 17408933.0,
|
|
"step": 8410
|
|
},
|
|
{
|
|
"entropy": 5.295473432540893,
|
|
"epoch": 7.643343934575193,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00012025700465091177,
|
|
"loss": 4.4766,
|
|
"mean_token_accuracy": 0.2493545889854431,
|
|
"num_tokens": 17419391.0,
|
|
"step": 8415
|
|
},
|
|
{
|
|
"entropy": 5.296148300170898,
|
|
"epoch": 7.647887323943662,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00012000062347482413,
|
|
"loss": 4.5342,
|
|
"mean_token_accuracy": 0.24442587196826934,
|
|
"num_tokens": 17430562.0,
|
|
"step": 8420
|
|
},
|
|
{
|
|
"entropy": 5.263896465301514,
|
|
"epoch": 7.652430713312131,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00011974462474429015,
|
|
"loss": 4.4695,
|
|
"mean_token_accuracy": 0.24573067277669908,
|
|
"num_tokens": 17441345.0,
|
|
"step": 8425
|
|
},
|
|
{
|
|
"entropy": 5.244579458236695,
|
|
"epoch": 7.6569741026806,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00011948900909096101,
|
|
"loss": 4.4927,
|
|
"mean_token_accuracy": 0.24498358070850373,
|
|
"num_tokens": 17450972.0,
|
|
"step": 8430
|
|
},
|
|
{
|
|
"entropy": 5.272073650360108,
|
|
"epoch": 7.661517492049069,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001192337771455431,
|
|
"loss": 4.5108,
|
|
"mean_token_accuracy": 0.24020439088344575,
|
|
"num_tokens": 17460930.0,
|
|
"step": 8435
|
|
},
|
|
{
|
|
"entropy": 5.326716661453247,
|
|
"epoch": 7.666060881417538,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00011897892953779576,
|
|
"loss": 4.4785,
|
|
"mean_token_accuracy": 0.24736214876174928,
|
|
"num_tokens": 17470865.0,
|
|
"step": 8440
|
|
},
|
|
{
|
|
"entropy": 5.271138381958008,
|
|
"epoch": 7.670604270786006,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00011872446689653032,
|
|
"loss": 4.4146,
|
|
"mean_token_accuracy": 0.25674081295728685,
|
|
"num_tokens": 17481379.0,
|
|
"step": 8445
|
|
},
|
|
{
|
|
"entropy": 5.381897401809693,
|
|
"epoch": 7.675147660154475,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0001184703898496078,
|
|
"loss": 4.5633,
|
|
"mean_token_accuracy": 0.243897944688797,
|
|
"num_tokens": 17492741.0,
|
|
"step": 8450
|
|
},
|
|
{
|
|
"entropy": 5.297904968261719,
|
|
"epoch": 7.679691049522944,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00011821669902393823,
|
|
"loss": 4.5511,
|
|
"mean_token_accuracy": 0.23675094097852706,
|
|
"num_tokens": 17504179.0,
|
|
"step": 8455
|
|
},
|
|
{
|
|
"entropy": 5.2538878440856935,
|
|
"epoch": 7.684234438891413,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00011796339504547846,
|
|
"loss": 4.4874,
|
|
"mean_token_accuracy": 0.2537557780742645,
|
|
"num_tokens": 17514299.0,
|
|
"step": 8460
|
|
},
|
|
{
|
|
"entropy": 5.322014570236206,
|
|
"epoch": 7.688777828259882,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00011771047853923086,
|
|
"loss": 4.4861,
|
|
"mean_token_accuracy": 0.25053404122591016,
|
|
"num_tokens": 17524454.0,
|
|
"step": 8465
|
|
},
|
|
{
|
|
"entropy": 5.310614776611328,
|
|
"epoch": 7.693321217628351,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00011745795012924169,
|
|
"loss": 4.4626,
|
|
"mean_token_accuracy": 0.2574162095785141,
|
|
"num_tokens": 17534887.0,
|
|
"step": 8470
|
|
},
|
|
{
|
|
"entropy": 5.262013339996338,
|
|
"epoch": 7.69786460699682,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00011720581043859988,
|
|
"loss": 4.5344,
|
|
"mean_token_accuracy": 0.24677640646696092,
|
|
"num_tokens": 17544944.0,
|
|
"step": 8475
|
|
},
|
|
{
|
|
"entropy": 5.28565993309021,
|
|
"epoch": 7.702407996365288,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00011695406008943495,
|
|
"loss": 4.4791,
|
|
"mean_token_accuracy": 0.25457853376865386,
|
|
"num_tokens": 17556481.0,
|
|
"step": 8480
|
|
},
|
|
{
|
|
"entropy": 5.3391670227050785,
|
|
"epoch": 7.706951385733757,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00011670269970291584,
|
|
"loss": 4.546,
|
|
"mean_token_accuracy": 0.2420789435505867,
|
|
"num_tokens": 17566776.0,
|
|
"step": 8485
|
|
},
|
|
{
|
|
"entropy": 5.26668701171875,
|
|
"epoch": 7.711494775102226,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001164517298992493,
|
|
"loss": 4.4518,
|
|
"mean_token_accuracy": 0.24811847358942032,
|
|
"num_tokens": 17575441.0,
|
|
"step": 8490
|
|
},
|
|
{
|
|
"entropy": 5.3087568283081055,
|
|
"epoch": 7.716038164470695,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00011620115129767848,
|
|
"loss": 4.5835,
|
|
"mean_token_accuracy": 0.24012621343135834,
|
|
"num_tokens": 17585792.0,
|
|
"step": 8495
|
|
},
|
|
{
|
|
"entropy": 5.270870494842529,
|
|
"epoch": 7.720581553839164,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00011595096451648111,
|
|
"loss": 4.5003,
|
|
"mean_token_accuracy": 0.24123722016811372,
|
|
"num_tokens": 17595820.0,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"epoch": 7.720581553839164,
|
|
"eval_entropy": 5.160217816330666,
|
|
"eval_loss": 5.668470859527588,
|
|
"eval_mean_token_accuracy": 0.17512425825346348,
|
|
"eval_num_tokens": 17595820.0,
|
|
"eval_runtime": 2.1838,
|
|
"eval_samples_per_second": 1572.916,
|
|
"eval_steps_per_second": 196.901,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"entropy": 5.275099420547486,
|
|
"epoch": 7.725124943207633,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00011570117017296819,
|
|
"loss": 4.4495,
|
|
"mean_token_accuracy": 0.24744274318218232,
|
|
"num_tokens": 17605791.0,
|
|
"step": 8505
|
|
},
|
|
{
|
|
"entropy": 5.277965068817139,
|
|
"epoch": 7.729668332576102,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00011545176888348237,
|
|
"loss": 4.509,
|
|
"mean_token_accuracy": 0.24276671260595323,
|
|
"num_tokens": 17616159.0,
|
|
"step": 8510
|
|
},
|
|
{
|
|
"entropy": 5.261576080322266,
|
|
"epoch": 7.734211721944571,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00011520276126339667,
|
|
"loss": 4.4769,
|
|
"mean_token_accuracy": 0.24602554440498353,
|
|
"num_tokens": 17626612.0,
|
|
"step": 8515
|
|
},
|
|
{
|
|
"entropy": 5.290240001678467,
|
|
"epoch": 7.73875511131304,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00011495414792711259,
|
|
"loss": 4.4896,
|
|
"mean_token_accuracy": 0.24499859064817428,
|
|
"num_tokens": 17636254.0,
|
|
"step": 8520
|
|
},
|
|
{
|
|
"entropy": 5.370635986328125,
|
|
"epoch": 7.743298500681508,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00011470592948805882,
|
|
"loss": 4.5868,
|
|
"mean_token_accuracy": 0.2344598650932312,
|
|
"num_tokens": 17648097.0,
|
|
"step": 8525
|
|
},
|
|
{
|
|
"entropy": 5.273475790023804,
|
|
"epoch": 7.747841890049977,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00011445810655868957,
|
|
"loss": 4.469,
|
|
"mean_token_accuracy": 0.25006858855485914,
|
|
"num_tokens": 17658361.0,
|
|
"step": 8530
|
|
},
|
|
{
|
|
"entropy": 5.210960626602173,
|
|
"epoch": 7.752385279418446,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00011421067975048355,
|
|
"loss": 4.4302,
|
|
"mean_token_accuracy": 0.2549549758434296,
|
|
"num_tokens": 17668640.0,
|
|
"step": 8535
|
|
},
|
|
{
|
|
"entropy": 5.222963571548462,
|
|
"epoch": 7.756928668786915,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011396364967394152,
|
|
"loss": 4.3787,
|
|
"mean_token_accuracy": 0.26243205219507215,
|
|
"num_tokens": 17678257.0,
|
|
"step": 8540
|
|
},
|
|
{
|
|
"entropy": 5.2381000995635985,
|
|
"epoch": 7.761472058155384,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00011371701693858582,
|
|
"loss": 4.4943,
|
|
"mean_token_accuracy": 0.24916886836290358,
|
|
"num_tokens": 17688635.0,
|
|
"step": 8545
|
|
},
|
|
{
|
|
"entropy": 5.296449375152588,
|
|
"epoch": 7.766015447523853,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00011347078215295817,
|
|
"loss": 4.5497,
|
|
"mean_token_accuracy": 0.23976572155952453,
|
|
"num_tokens": 17699917.0,
|
|
"step": 8550
|
|
},
|
|
{
|
|
"entropy": 5.31504111289978,
|
|
"epoch": 7.770558836892322,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00011322494592461839,
|
|
"loss": 4.4972,
|
|
"mean_token_accuracy": 0.25059440433979036,
|
|
"num_tokens": 17710278.0,
|
|
"step": 8555
|
|
},
|
|
{
|
|
"entropy": 5.265868234634399,
|
|
"epoch": 7.77510222626079,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00011297950886014292,
|
|
"loss": 4.4763,
|
|
"mean_token_accuracy": 0.2534422755241394,
|
|
"num_tokens": 17720375.0,
|
|
"step": 8560
|
|
},
|
|
{
|
|
"entropy": 5.25705394744873,
|
|
"epoch": 7.779645615629259,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001127344715651234,
|
|
"loss": 4.5176,
|
|
"mean_token_accuracy": 0.23520173132419586,
|
|
"num_tokens": 17730299.0,
|
|
"step": 8565
|
|
},
|
|
{
|
|
"entropy": 5.306503343582153,
|
|
"epoch": 7.784189004997728,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00011248983464416497,
|
|
"loss": 4.5241,
|
|
"mean_token_accuracy": 0.24684132188558577,
|
|
"num_tokens": 17741096.0,
|
|
"step": 8570
|
|
},
|
|
{
|
|
"entropy": 5.250234365463257,
|
|
"epoch": 7.788732394366197,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00011224559870088488,
|
|
"loss": 4.4836,
|
|
"mean_token_accuracy": 0.24645259827375413,
|
|
"num_tokens": 17751502.0,
|
|
"step": 8575
|
|
},
|
|
{
|
|
"entropy": 5.280009460449219,
|
|
"epoch": 7.793275783734666,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00011200176433791101,
|
|
"loss": 4.5084,
|
|
"mean_token_accuracy": 0.25343816727399826,
|
|
"num_tokens": 17761955.0,
|
|
"step": 8580
|
|
},
|
|
{
|
|
"entropy": 5.249505043029785,
|
|
"epoch": 7.797819173103135,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001117583321568805,
|
|
"loss": 4.4655,
|
|
"mean_token_accuracy": 0.24491624683141708,
|
|
"num_tokens": 17772330.0,
|
|
"step": 8585
|
|
},
|
|
{
|
|
"entropy": 5.286121463775634,
|
|
"epoch": 7.802362562471604,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00011151530275843805,
|
|
"loss": 4.4249,
|
|
"mean_token_accuracy": 0.2506417602300644,
|
|
"num_tokens": 17781688.0,
|
|
"step": 8590
|
|
},
|
|
{
|
|
"entropy": 5.245406246185302,
|
|
"epoch": 7.806905951840073,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011127267674223449,
|
|
"loss": 4.458,
|
|
"mean_token_accuracy": 0.25112144947052,
|
|
"num_tokens": 17792037.0,
|
|
"step": 8595
|
|
},
|
|
{
|
|
"entropy": 5.284047555923462,
|
|
"epoch": 7.811449341208542,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00011103045470692532,
|
|
"loss": 4.5458,
|
|
"mean_token_accuracy": 0.2463863968849182,
|
|
"num_tokens": 17802705.0,
|
|
"step": 8600
|
|
},
|
|
{
|
|
"entropy": 5.344259405136109,
|
|
"epoch": 7.81599273057701,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00011078863725016955,
|
|
"loss": 4.5915,
|
|
"mean_token_accuracy": 0.23911388516426085,
|
|
"num_tokens": 17813758.0,
|
|
"step": 8605
|
|
},
|
|
{
|
|
"entropy": 5.250797271728516,
|
|
"epoch": 7.820536119945479,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011054722496862759,
|
|
"loss": 4.4636,
|
|
"mean_token_accuracy": 0.25031974762678144,
|
|
"num_tokens": 17823465.0,
|
|
"step": 8610
|
|
},
|
|
{
|
|
"entropy": 5.2879729747772215,
|
|
"epoch": 7.825079509313948,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00011030621845796021,
|
|
"loss": 4.4874,
|
|
"mean_token_accuracy": 0.2505660682916641,
|
|
"num_tokens": 17833719.0,
|
|
"step": 8615
|
|
},
|
|
{
|
|
"entropy": 5.261925268173218,
|
|
"epoch": 7.829622898682417,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00011006561831282703,
|
|
"loss": 4.4589,
|
|
"mean_token_accuracy": 0.2567958801984787,
|
|
"num_tokens": 17844519.0,
|
|
"step": 8620
|
|
},
|
|
{
|
|
"entropy": 5.25709867477417,
|
|
"epoch": 7.834166288050886,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010982542512688515,
|
|
"loss": 4.4553,
|
|
"mean_token_accuracy": 0.2526007011532784,
|
|
"num_tokens": 17856463.0,
|
|
"step": 8625
|
|
},
|
|
{
|
|
"entropy": 5.221201229095459,
|
|
"epoch": 7.838709677419355,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00010958563949278711,
|
|
"loss": 4.4524,
|
|
"mean_token_accuracy": 0.24717678874731064,
|
|
"num_tokens": 17867449.0,
|
|
"step": 8630
|
|
},
|
|
{
|
|
"entropy": 5.202815008163452,
|
|
"epoch": 7.843253066787824,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00010934626200218032,
|
|
"loss": 4.4601,
|
|
"mean_token_accuracy": 0.2472052425146103,
|
|
"num_tokens": 17878522.0,
|
|
"step": 8635
|
|
},
|
|
{
|
|
"entropy": 5.302530527114868,
|
|
"epoch": 7.847796456156293,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00010910729324570488,
|
|
"loss": 4.5313,
|
|
"mean_token_accuracy": 0.24312609136104585,
|
|
"num_tokens": 17888544.0,
|
|
"step": 8640
|
|
},
|
|
{
|
|
"entropy": 5.3207478523254395,
|
|
"epoch": 7.852339845524762,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00010886873381299248,
|
|
"loss": 4.5709,
|
|
"mean_token_accuracy": 0.2403368279337883,
|
|
"num_tokens": 17897432.0,
|
|
"step": 8645
|
|
},
|
|
{
|
|
"entropy": 5.284350109100342,
|
|
"epoch": 7.85688323489323,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00010863058429266469,
|
|
"loss": 4.5041,
|
|
"mean_token_accuracy": 0.2518989682197571,
|
|
"num_tokens": 17907868.0,
|
|
"step": 8650
|
|
},
|
|
{
|
|
"entropy": 5.305052804946899,
|
|
"epoch": 7.861426624261699,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00010839284527233192,
|
|
"loss": 4.454,
|
|
"mean_token_accuracy": 0.24648875296115874,
|
|
"num_tokens": 17917887.0,
|
|
"step": 8655
|
|
},
|
|
{
|
|
"entropy": 5.2419274806976315,
|
|
"epoch": 7.865970013630168,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00010815551733859153,
|
|
"loss": 4.4509,
|
|
"mean_token_accuracy": 0.24870684295892714,
|
|
"num_tokens": 17928030.0,
|
|
"step": 8660
|
|
},
|
|
{
|
|
"entropy": 5.260346698760986,
|
|
"epoch": 7.870513402998637,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00010791860107702658,
|
|
"loss": 4.5753,
|
|
"mean_token_accuracy": 0.2352682128548622,
|
|
"num_tokens": 17937990.0,
|
|
"step": 8665
|
|
},
|
|
{
|
|
"entropy": 5.251736736297607,
|
|
"epoch": 7.875056792367106,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010768209707220435,
|
|
"loss": 4.555,
|
|
"mean_token_accuracy": 0.2428753674030304,
|
|
"num_tokens": 17948262.0,
|
|
"step": 8670
|
|
},
|
|
{
|
|
"entropy": 5.289690113067627,
|
|
"epoch": 7.879600181735575,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00010744600590767504,
|
|
"loss": 4.5308,
|
|
"mean_token_accuracy": 0.24526872038841246,
|
|
"num_tokens": 17958942.0,
|
|
"step": 8675
|
|
},
|
|
{
|
|
"entropy": 5.259433460235596,
|
|
"epoch": 7.884143571104044,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010721032816597012,
|
|
"loss": 4.4982,
|
|
"mean_token_accuracy": 0.24745635837316513,
|
|
"num_tokens": 17969362.0,
|
|
"step": 8680
|
|
},
|
|
{
|
|
"entropy": 5.304601240158081,
|
|
"epoch": 7.888686960472512,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000106975064428601,
|
|
"loss": 4.4857,
|
|
"mean_token_accuracy": 0.24950945377349854,
|
|
"num_tokens": 17979009.0,
|
|
"step": 8685
|
|
},
|
|
{
|
|
"entropy": 5.318955898284912,
|
|
"epoch": 7.893230349840981,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00010674021527605742,
|
|
"loss": 4.5312,
|
|
"mean_token_accuracy": 0.2472262352705002,
|
|
"num_tokens": 17989064.0,
|
|
"step": 8690
|
|
},
|
|
{
|
|
"entropy": 5.247525882720947,
|
|
"epoch": 7.89777373920945,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00010650578128780656,
|
|
"loss": 4.467,
|
|
"mean_token_accuracy": 0.25447062999010084,
|
|
"num_tokens": 17998574.0,
|
|
"step": 8695
|
|
},
|
|
{
|
|
"entropy": 5.356578922271728,
|
|
"epoch": 7.902317128577919,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001062717630422908,
|
|
"loss": 4.6644,
|
|
"mean_token_accuracy": 0.23843247294425965,
|
|
"num_tokens": 18010735.0,
|
|
"step": 8700
|
|
},
|
|
{
|
|
"entropy": 5.278338956832886,
|
|
"epoch": 7.906860517946388,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00010603816111692699,
|
|
"loss": 4.4936,
|
|
"mean_token_accuracy": 0.2506414696574211,
|
|
"num_tokens": 18020316.0,
|
|
"step": 8705
|
|
},
|
|
{
|
|
"entropy": 5.299891471862793,
|
|
"epoch": 7.911403907314857,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00010580497608810455,
|
|
"loss": 4.4961,
|
|
"mean_token_accuracy": 0.24570607841014863,
|
|
"num_tokens": 18031058.0,
|
|
"step": 8710
|
|
},
|
|
{
|
|
"entropy": 5.271481561660766,
|
|
"epoch": 7.915947296683326,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00010557220853118456,
|
|
"loss": 4.4872,
|
|
"mean_token_accuracy": 0.25166180580854414,
|
|
"num_tokens": 18040603.0,
|
|
"step": 8715
|
|
},
|
|
{
|
|
"entropy": 5.335350418090821,
|
|
"epoch": 7.920490686051795,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00010533985902049755,
|
|
"loss": 4.5654,
|
|
"mean_token_accuracy": 0.24333755373954774,
|
|
"num_tokens": 18051369.0,
|
|
"step": 8720
|
|
},
|
|
{
|
|
"entropy": 5.261841297149658,
|
|
"epoch": 7.925034075420264,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010510792812934307,
|
|
"loss": 4.4539,
|
|
"mean_token_accuracy": 0.25583341121673586,
|
|
"num_tokens": 18062283.0,
|
|
"step": 8725
|
|
},
|
|
{
|
|
"entropy": 5.302979850769043,
|
|
"epoch": 7.929577464788732,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00010487641642998746,
|
|
"loss": 4.5203,
|
|
"mean_token_accuracy": 0.24689577966928483,
|
|
"num_tokens": 18072125.0,
|
|
"step": 8730
|
|
},
|
|
{
|
|
"entropy": 5.27326135635376,
|
|
"epoch": 7.934120854157201,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00010464532449366281,
|
|
"loss": 4.5558,
|
|
"mean_token_accuracy": 0.23758892416954042,
|
|
"num_tokens": 18083812.0,
|
|
"step": 8735
|
|
},
|
|
{
|
|
"entropy": 5.307485151290893,
|
|
"epoch": 7.93866424352567,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00010441465289056548,
|
|
"loss": 4.5296,
|
|
"mean_token_accuracy": 0.24709307700395583,
|
|
"num_tokens": 18095399.0,
|
|
"step": 8740
|
|
},
|
|
{
|
|
"entropy": 5.304354619979859,
|
|
"epoch": 7.943207632894139,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00010418440218985482,
|
|
"loss": 4.5218,
|
|
"mean_token_accuracy": 0.25186712890863416,
|
|
"num_tokens": 18106234.0,
|
|
"step": 8745
|
|
},
|
|
{
|
|
"entropy": 5.25816764831543,
|
|
"epoch": 7.947751022262608,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00010395457295965143,
|
|
"loss": 4.4211,
|
|
"mean_token_accuracy": 0.2605363577604294,
|
|
"num_tokens": 18116129.0,
|
|
"step": 8750
|
|
},
|
|
{
|
|
"entropy": 5.340640020370484,
|
|
"epoch": 7.952294411631077,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010372516576703617,
|
|
"loss": 4.55,
|
|
"mean_token_accuracy": 0.24606338143348694,
|
|
"num_tokens": 18126757.0,
|
|
"step": 8755
|
|
},
|
|
{
|
|
"entropy": 5.270389223098755,
|
|
"epoch": 7.956837800999546,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001034961811780484,
|
|
"loss": 4.511,
|
|
"mean_token_accuracy": 0.2438106894493103,
|
|
"num_tokens": 18136645.0,
|
|
"step": 8760
|
|
},
|
|
{
|
|
"entropy": 5.329086637496948,
|
|
"epoch": 7.961381190368014,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00010326761975768492,
|
|
"loss": 4.5673,
|
|
"mean_token_accuracy": 0.24086528718471528,
|
|
"num_tokens": 18146458.0,
|
|
"step": 8765
|
|
},
|
|
{
|
|
"entropy": 5.296896982192993,
|
|
"epoch": 7.965924579736483,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00010303948206989827,
|
|
"loss": 4.4968,
|
|
"mean_token_accuracy": 0.25017675161361697,
|
|
"num_tokens": 18156151.0,
|
|
"step": 8770
|
|
},
|
|
{
|
|
"entropy": 5.282786893844604,
|
|
"epoch": 7.970467969104952,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010281176867759556,
|
|
"loss": 4.5256,
|
|
"mean_token_accuracy": 0.2477202445268631,
|
|
"num_tokens": 18166695.0,
|
|
"step": 8775
|
|
},
|
|
{
|
|
"entropy": 5.262575721740722,
|
|
"epoch": 7.975011358473421,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010258448014263682,
|
|
"loss": 4.4726,
|
|
"mean_token_accuracy": 0.2501365527510643,
|
|
"num_tokens": 18177011.0,
|
|
"step": 8780
|
|
},
|
|
{
|
|
"entropy": 5.30100793838501,
|
|
"epoch": 7.97955474784189,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010235761702583405,
|
|
"loss": 4.5741,
|
|
"mean_token_accuracy": 0.2434493198990822,
|
|
"num_tokens": 18187179.0,
|
|
"step": 8785
|
|
},
|
|
{
|
|
"entropy": 5.272710847854614,
|
|
"epoch": 7.984098137210359,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010213117988694944,
|
|
"loss": 4.5092,
|
|
"mean_token_accuracy": 0.24266156405210496,
|
|
"num_tokens": 18197106.0,
|
|
"step": 8790
|
|
},
|
|
{
|
|
"entropy": 5.206980895996094,
|
|
"epoch": 7.988641526578828,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00010190516928469407,
|
|
"loss": 4.4982,
|
|
"mean_token_accuracy": 0.2459101766347885,
|
|
"num_tokens": 18207729.0,
|
|
"step": 8795
|
|
},
|
|
{
|
|
"entropy": 5.266595029830933,
|
|
"epoch": 7.993184915947297,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00010167958577672654,
|
|
"loss": 4.4526,
|
|
"mean_token_accuracy": 0.24664996564388275,
|
|
"num_tokens": 18217885.0,
|
|
"step": 8800
|
|
},
|
|
{
|
|
"entropy": 5.300826597213745,
|
|
"epoch": 7.997728305315766,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00010145442991965198,
|
|
"loss": 4.5421,
|
|
"mean_token_accuracy": 0.24420285522937774,
|
|
"num_tokens": 18229633.0,
|
|
"step": 8805
|
|
},
|
|
{
|
|
"entropy": 5.241123729281956,
|
|
"epoch": 8.001817355747388,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00010122970226901981,
|
|
"loss": 4.4969,
|
|
"mean_token_accuracy": 0.2460640917221705,
|
|
"num_tokens": 18239194.0,
|
|
"step": 8810
|
|
},
|
|
{
|
|
"entropy": 5.313464498519897,
|
|
"epoch": 8.006360745115856,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00010100540337932339,
|
|
"loss": 4.4751,
|
|
"mean_token_accuracy": 0.24663930684328078,
|
|
"num_tokens": 18250061.0,
|
|
"step": 8815
|
|
},
|
|
{
|
|
"entropy": 5.284324502944946,
|
|
"epoch": 8.010904134484326,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00010078153380399773,
|
|
"loss": 4.3935,
|
|
"mean_token_accuracy": 0.2654407650232315,
|
|
"num_tokens": 18260977.0,
|
|
"step": 8820
|
|
},
|
|
{
|
|
"entropy": 5.279210758209229,
|
|
"epoch": 8.015447523852794,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010055809409541902,
|
|
"loss": 4.4461,
|
|
"mean_token_accuracy": 0.2541866168379784,
|
|
"num_tokens": 18271866.0,
|
|
"step": 8825
|
|
},
|
|
{
|
|
"entropy": 5.247068452835083,
|
|
"epoch": 8.019990913221264,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00010033508480490222,
|
|
"loss": 4.495,
|
|
"mean_token_accuracy": 0.24897793233394622,
|
|
"num_tokens": 18283353.0,
|
|
"step": 8830
|
|
},
|
|
{
|
|
"entropy": 5.208508443832398,
|
|
"epoch": 8.024534302589732,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00010011250648270081,
|
|
"loss": 4.4297,
|
|
"mean_token_accuracy": 0.25143207907676696,
|
|
"num_tokens": 18293619.0,
|
|
"step": 8835
|
|
},
|
|
{
|
|
"entropy": 5.307828187942505,
|
|
"epoch": 8.0290776919582,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.98903596780046e-05,
|
|
"loss": 4.4262,
|
|
"mean_token_accuracy": 0.2614478632807732,
|
|
"num_tokens": 18303411.0,
|
|
"step": 8840
|
|
},
|
|
{
|
|
"entropy": 5.264690637588501,
|
|
"epoch": 8.03362108132667,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.966864493893876e-05,
|
|
"loss": 4.3854,
|
|
"mean_token_accuracy": 0.2620695993304253,
|
|
"num_tokens": 18312798.0,
|
|
"step": 8845
|
|
},
|
|
{
|
|
"entropy": 5.245627450942993,
|
|
"epoch": 8.038164470695138,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.94473628125623e-05,
|
|
"loss": 4.4517,
|
|
"mean_token_accuracy": 0.24819066673517226,
|
|
"num_tokens": 18324111.0,
|
|
"step": 8850
|
|
},
|
|
{
|
|
"entropy": 5.254646921157837,
|
|
"epoch": 8.042707860063608,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.922651384486695e-05,
|
|
"loss": 4.4191,
|
|
"mean_token_accuracy": 0.24866719543933868,
|
|
"num_tokens": 18335197.0,
|
|
"step": 8855
|
|
},
|
|
{
|
|
"entropy": 5.309178638458252,
|
|
"epoch": 8.047251249432076,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 9.900609858077562e-05,
|
|
"loss": 4.4805,
|
|
"mean_token_accuracy": 0.24895586371421813,
|
|
"num_tokens": 18345896.0,
|
|
"step": 8860
|
|
},
|
|
{
|
|
"entropy": 5.3404608249664305,
|
|
"epoch": 8.051794638800546,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 9.878611756414094e-05,
|
|
"loss": 4.4996,
|
|
"mean_token_accuracy": 0.24642963111400604,
|
|
"num_tokens": 18356941.0,
|
|
"step": 8865
|
|
},
|
|
{
|
|
"entropy": 5.2591893672943115,
|
|
"epoch": 8.056338028169014,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 9.856657133774424e-05,
|
|
"loss": 4.3878,
|
|
"mean_token_accuracy": 0.2581391528248787,
|
|
"num_tokens": 18366767.0,
|
|
"step": 8870
|
|
},
|
|
{
|
|
"entropy": 5.311511850357055,
|
|
"epoch": 8.060881417537482,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.834746044329411e-05,
|
|
"loss": 4.5115,
|
|
"mean_token_accuracy": 0.25048388093709945,
|
|
"num_tokens": 18376698.0,
|
|
"step": 8875
|
|
},
|
|
{
|
|
"entropy": 5.255208301544189,
|
|
"epoch": 8.065424806905952,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.81287854214248e-05,
|
|
"loss": 4.4163,
|
|
"mean_token_accuracy": 0.2519300624728203,
|
|
"num_tokens": 18386678.0,
|
|
"step": 8880
|
|
},
|
|
{
|
|
"entropy": 5.281863594055176,
|
|
"epoch": 8.06996819627442,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 9.791054681169522e-05,
|
|
"loss": 4.4439,
|
|
"mean_token_accuracy": 0.2537225350737572,
|
|
"num_tokens": 18397020.0,
|
|
"step": 8885
|
|
},
|
|
{
|
|
"entropy": 5.229952478408814,
|
|
"epoch": 8.07451158564289,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.769274515258739e-05,
|
|
"loss": 4.4221,
|
|
"mean_token_accuracy": 0.24786950498819352,
|
|
"num_tokens": 18407330.0,
|
|
"step": 8890
|
|
},
|
|
{
|
|
"entropy": 5.2420032024383545,
|
|
"epoch": 8.079054975011358,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 9.747538098150539e-05,
|
|
"loss": 4.4346,
|
|
"mean_token_accuracy": 0.25157456248998644,
|
|
"num_tokens": 18417046.0,
|
|
"step": 8895
|
|
},
|
|
{
|
|
"entropy": 5.282014036178589,
|
|
"epoch": 8.083598364379828,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 9.725845483477352e-05,
|
|
"loss": 4.4832,
|
|
"mean_token_accuracy": 0.25494079738855363,
|
|
"num_tokens": 18427907.0,
|
|
"step": 8900
|
|
},
|
|
{
|
|
"entropy": 5.260359144210815,
|
|
"epoch": 8.088141753748296,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 9.704196724763565e-05,
|
|
"loss": 4.4388,
|
|
"mean_token_accuracy": 0.2513161689043045,
|
|
"num_tokens": 18437972.0,
|
|
"step": 8905
|
|
},
|
|
{
|
|
"entropy": 5.241520023345947,
|
|
"epoch": 8.092685143116766,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.682591875425323e-05,
|
|
"loss": 4.455,
|
|
"mean_token_accuracy": 0.25630090981721876,
|
|
"num_tokens": 18448188.0,
|
|
"step": 8910
|
|
},
|
|
{
|
|
"entropy": 5.268462181091309,
|
|
"epoch": 8.097228532485234,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 9.661030988770463e-05,
|
|
"loss": 4.4711,
|
|
"mean_token_accuracy": 0.2498174026608467,
|
|
"num_tokens": 18458130.0,
|
|
"step": 8915
|
|
},
|
|
{
|
|
"entropy": 5.280180025100708,
|
|
"epoch": 8.101771921853702,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.63951411799831e-05,
|
|
"loss": 4.4359,
|
|
"mean_token_accuracy": 0.2548591628670692,
|
|
"num_tokens": 18468257.0,
|
|
"step": 8920
|
|
},
|
|
{
|
|
"entropy": 5.241324043273925,
|
|
"epoch": 8.106315311222172,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 9.618041316199617e-05,
|
|
"loss": 4.4585,
|
|
"mean_token_accuracy": 0.24870146214962005,
|
|
"num_tokens": 18478720.0,
|
|
"step": 8925
|
|
},
|
|
{
|
|
"entropy": 5.290587520599365,
|
|
"epoch": 8.11085870059064,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 9.596612636356383e-05,
|
|
"loss": 4.4354,
|
|
"mean_token_accuracy": 0.2539619356393814,
|
|
"num_tokens": 18489543.0,
|
|
"step": 8930
|
|
},
|
|
{
|
|
"entropy": 5.278292083740235,
|
|
"epoch": 8.11540208995911,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.575228131341745e-05,
|
|
"loss": 4.4637,
|
|
"mean_token_accuracy": 0.251634855568409,
|
|
"num_tokens": 18499505.0,
|
|
"step": 8935
|
|
},
|
|
{
|
|
"entropy": 5.325848913192749,
|
|
"epoch": 8.119945479327578,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.55388785391984e-05,
|
|
"loss": 4.5274,
|
|
"mean_token_accuracy": 0.2469608426094055,
|
|
"num_tokens": 18510044.0,
|
|
"step": 8940
|
|
},
|
|
{
|
|
"entropy": 5.282946634292602,
|
|
"epoch": 8.124488868696048,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 9.532591856745689e-05,
|
|
"loss": 4.4617,
|
|
"mean_token_accuracy": 0.25159019976854324,
|
|
"num_tokens": 18520537.0,
|
|
"step": 8945
|
|
},
|
|
{
|
|
"entropy": 5.281657075881958,
|
|
"epoch": 8.129032258064516,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 9.511340192365044e-05,
|
|
"loss": 4.5134,
|
|
"mean_token_accuracy": 0.2421749636530876,
|
|
"num_tokens": 18531840.0,
|
|
"step": 8950
|
|
},
|
|
{
|
|
"entropy": 5.260400390625,
|
|
"epoch": 8.133575647432984,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 9.490132913214277e-05,
|
|
"loss": 4.4135,
|
|
"mean_token_accuracy": 0.25366479009389875,
|
|
"num_tokens": 18542097.0,
|
|
"step": 8955
|
|
},
|
|
{
|
|
"entropy": 5.288394927978516,
|
|
"epoch": 8.138119036801454,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.468970071620231e-05,
|
|
"loss": 4.4628,
|
|
"mean_token_accuracy": 0.2525705724954605,
|
|
"num_tokens": 18551629.0,
|
|
"step": 8960
|
|
},
|
|
{
|
|
"entropy": 5.243550634384155,
|
|
"epoch": 8.142662426169922,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 9.447851719800126e-05,
|
|
"loss": 4.4091,
|
|
"mean_token_accuracy": 0.25359111130237577,
|
|
"num_tokens": 18562248.0,
|
|
"step": 8965
|
|
},
|
|
{
|
|
"entropy": 5.246295928955078,
|
|
"epoch": 8.147205815538392,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 9.426777909861395e-05,
|
|
"loss": 4.4068,
|
|
"mean_token_accuracy": 0.2578919529914856,
|
|
"num_tokens": 18572107.0,
|
|
"step": 8970
|
|
},
|
|
{
|
|
"entropy": 5.267754125595093,
|
|
"epoch": 8.15174920490686,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 9.405748693801562e-05,
|
|
"loss": 4.4883,
|
|
"mean_token_accuracy": 0.2444618374109268,
|
|
"num_tokens": 18582665.0,
|
|
"step": 8975
|
|
},
|
|
{
|
|
"entropy": 5.355312061309815,
|
|
"epoch": 8.15629259427533,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.384764123508124e-05,
|
|
"loss": 4.4952,
|
|
"mean_token_accuracy": 0.24852756559848785,
|
|
"num_tokens": 18593737.0,
|
|
"step": 8980
|
|
},
|
|
{
|
|
"entropy": 5.292559719085693,
|
|
"epoch": 8.160835983643798,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 9.363824250758443e-05,
|
|
"loss": 4.4529,
|
|
"mean_token_accuracy": 0.2584801286458969,
|
|
"num_tokens": 18604180.0,
|
|
"step": 8985
|
|
},
|
|
{
|
|
"entropy": 5.222184705734253,
|
|
"epoch": 8.165379373012268,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 9.34292912721955e-05,
|
|
"loss": 4.3828,
|
|
"mean_token_accuracy": 0.2555510401725769,
|
|
"num_tokens": 18614328.0,
|
|
"step": 8990
|
|
},
|
|
{
|
|
"entropy": 5.287621593475341,
|
|
"epoch": 8.169922762380736,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 9.322078804448095e-05,
|
|
"loss": 4.4358,
|
|
"mean_token_accuracy": 0.26819380670785903,
|
|
"num_tokens": 18625357.0,
|
|
"step": 8995
|
|
},
|
|
{
|
|
"entropy": 5.23282265663147,
|
|
"epoch": 8.174466151749204,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 9.301273333890171e-05,
|
|
"loss": 4.4083,
|
|
"mean_token_accuracy": 0.25698724389076233,
|
|
"num_tokens": 18636306.0,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"epoch": 8.174466151749204,
|
|
"eval_entropy": 5.147978244825851,
|
|
"eval_loss": 5.677839756011963,
|
|
"eval_mean_token_accuracy": 0.17471307907686678,
|
|
"eval_num_tokens": 18636306.0,
|
|
"eval_runtime": 2.0063,
|
|
"eval_samples_per_second": 1712.097,
|
|
"eval_steps_per_second": 214.324,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"entropy": 5.313176155090332,
|
|
"epoch": 8.179009541117674,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.280512766881227e-05,
|
|
"loss": 4.4678,
|
|
"mean_token_accuracy": 0.2558841273188591,
|
|
"num_tokens": 18647134.0,
|
|
"step": 9005
|
|
},
|
|
{
|
|
"entropy": 5.243971538543701,
|
|
"epoch": 8.183552930486142,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 9.25979715464588e-05,
|
|
"loss": 4.4026,
|
|
"mean_token_accuracy": 0.2581817671656609,
|
|
"num_tokens": 18656948.0,
|
|
"step": 9010
|
|
},
|
|
{
|
|
"entropy": 5.271985769271851,
|
|
"epoch": 8.188096319854612,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.23912654829784e-05,
|
|
"loss": 4.4401,
|
|
"mean_token_accuracy": 0.24708039611577987,
|
|
"num_tokens": 18667682.0,
|
|
"step": 9015
|
|
},
|
|
{
|
|
"entropy": 5.289959144592285,
|
|
"epoch": 8.19263970922308,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 9.218500998839794e-05,
|
|
"loss": 4.5033,
|
|
"mean_token_accuracy": 0.25170744508504866,
|
|
"num_tokens": 18678124.0,
|
|
"step": 9020
|
|
},
|
|
{
|
|
"entropy": 5.270580625534057,
|
|
"epoch": 8.19718309859155,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.197920557163215e-05,
|
|
"loss": 4.4372,
|
|
"mean_token_accuracy": 0.2562348961830139,
|
|
"num_tokens": 18689063.0,
|
|
"step": 9025
|
|
},
|
|
{
|
|
"entropy": 5.300122547149658,
|
|
"epoch": 8.201726487960018,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 9.177385274048312e-05,
|
|
"loss": 4.4982,
|
|
"mean_token_accuracy": 0.25056360214948653,
|
|
"num_tokens": 18700118.0,
|
|
"step": 9030
|
|
},
|
|
{
|
|
"entropy": 5.306892538070679,
|
|
"epoch": 8.206269877328488,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 9.156895200163837e-05,
|
|
"loss": 4.4613,
|
|
"mean_token_accuracy": 0.2480306401848793,
|
|
"num_tokens": 18711683.0,
|
|
"step": 9035
|
|
},
|
|
{
|
|
"entropy": 5.293663120269775,
|
|
"epoch": 8.210813266696956,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.13645038606703e-05,
|
|
"loss": 4.4518,
|
|
"mean_token_accuracy": 0.24912430346012115,
|
|
"num_tokens": 18722033.0,
|
|
"step": 9040
|
|
},
|
|
{
|
|
"entropy": 5.31216983795166,
|
|
"epoch": 8.215356656065424,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.116050882203427e-05,
|
|
"loss": 4.4654,
|
|
"mean_token_accuracy": 0.24788122475147248,
|
|
"num_tokens": 18732387.0,
|
|
"step": 9045
|
|
},
|
|
{
|
|
"entropy": 5.312847709655761,
|
|
"epoch": 8.219900045433894,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.09569673890678e-05,
|
|
"loss": 4.4722,
|
|
"mean_token_accuracy": 0.25004308074712756,
|
|
"num_tokens": 18743573.0,
|
|
"step": 9050
|
|
},
|
|
{
|
|
"entropy": 5.1968591690063475,
|
|
"epoch": 8.224443434802362,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 9.075388006398907e-05,
|
|
"loss": 4.37,
|
|
"mean_token_accuracy": 0.2524448424577713,
|
|
"num_tokens": 18753897.0,
|
|
"step": 9055
|
|
},
|
|
{
|
|
"entropy": 5.288205575942993,
|
|
"epoch": 8.228986824170832,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.0551247347896e-05,
|
|
"loss": 4.4974,
|
|
"mean_token_accuracy": 0.24643170982599258,
|
|
"num_tokens": 18764849.0,
|
|
"step": 9060
|
|
},
|
|
{
|
|
"entropy": 5.236581707000733,
|
|
"epoch": 8.2335302135393,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.034906974076449e-05,
|
|
"loss": 4.3712,
|
|
"mean_token_accuracy": 0.2565824300050735,
|
|
"num_tokens": 18776404.0,
|
|
"step": 9065
|
|
},
|
|
{
|
|
"entropy": 5.287790679931641,
|
|
"epoch": 8.23807360290777,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.014734774144784e-05,
|
|
"loss": 4.4642,
|
|
"mean_token_accuracy": 0.24935608506202697,
|
|
"num_tokens": 18786803.0,
|
|
"step": 9070
|
|
},
|
|
{
|
|
"entropy": 5.289120817184449,
|
|
"epoch": 8.242616992276238,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 8.994608184767491e-05,
|
|
"loss": 4.4911,
|
|
"mean_token_accuracy": 0.24874145537614822,
|
|
"num_tokens": 18797135.0,
|
|
"step": 9075
|
|
},
|
|
{
|
|
"entropy": 5.280733585357666,
|
|
"epoch": 8.247160381644706,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 8.97452725560495e-05,
|
|
"loss": 4.3837,
|
|
"mean_token_accuracy": 0.2611017793416977,
|
|
"num_tokens": 18806495.0,
|
|
"step": 9080
|
|
},
|
|
{
|
|
"entropy": 5.284330987930298,
|
|
"epoch": 8.251703771013176,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.954492036204829e-05,
|
|
"loss": 4.4488,
|
|
"mean_token_accuracy": 0.2485547259449959,
|
|
"num_tokens": 18816925.0,
|
|
"step": 9085
|
|
},
|
|
{
|
|
"entropy": 5.285844039916992,
|
|
"epoch": 8.256247160381644,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 8.934502576002059e-05,
|
|
"loss": 4.4001,
|
|
"mean_token_accuracy": 0.25961398929357526,
|
|
"num_tokens": 18826604.0,
|
|
"step": 9090
|
|
},
|
|
{
|
|
"entropy": 5.298125314712524,
|
|
"epoch": 8.260790549750114,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.914558924318643e-05,
|
|
"loss": 4.4749,
|
|
"mean_token_accuracy": 0.2566651076078415,
|
|
"num_tokens": 18837278.0,
|
|
"step": 9095
|
|
},
|
|
{
|
|
"entropy": 5.273393011093139,
|
|
"epoch": 8.265333939118582,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 8.894661130363557e-05,
|
|
"loss": 4.4957,
|
|
"mean_token_accuracy": 0.2482885256409645,
|
|
"num_tokens": 18846588.0,
|
|
"step": 9100
|
|
},
|
|
{
|
|
"entropy": 5.227906227111816,
|
|
"epoch": 8.269877328487052,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 8.874809243232624e-05,
|
|
"loss": 4.4291,
|
|
"mean_token_accuracy": 0.2652652457356453,
|
|
"num_tokens": 18856345.0,
|
|
"step": 9105
|
|
},
|
|
{
|
|
"entropy": 5.284674882888794,
|
|
"epoch": 8.27442071785552,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.85500331190841e-05,
|
|
"loss": 4.5229,
|
|
"mean_token_accuracy": 0.2376490592956543,
|
|
"num_tokens": 18867000.0,
|
|
"step": 9110
|
|
},
|
|
{
|
|
"entropy": 5.25544958114624,
|
|
"epoch": 8.27896410722399,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 8.835243385260083e-05,
|
|
"loss": 4.4958,
|
|
"mean_token_accuracy": 0.24804212749004365,
|
|
"num_tokens": 18876925.0,
|
|
"step": 9115
|
|
},
|
|
{
|
|
"entropy": 5.303364038467407,
|
|
"epoch": 8.283507496592458,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 8.815529512043293e-05,
|
|
"loss": 4.4359,
|
|
"mean_token_accuracy": 0.2566846862435341,
|
|
"num_tokens": 18886376.0,
|
|
"step": 9120
|
|
},
|
|
{
|
|
"entropy": 5.271303319931031,
|
|
"epoch": 8.288050885960926,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.795861740900061e-05,
|
|
"loss": 4.4484,
|
|
"mean_token_accuracy": 0.25560179352760315,
|
|
"num_tokens": 18897258.0,
|
|
"step": 9125
|
|
},
|
|
{
|
|
"entropy": 5.317284870147705,
|
|
"epoch": 8.292594275329396,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 8.77624012035866e-05,
|
|
"loss": 4.4284,
|
|
"mean_token_accuracy": 0.26159475147724154,
|
|
"num_tokens": 18906959.0,
|
|
"step": 9130
|
|
},
|
|
{
|
|
"entropy": 5.290068769454956,
|
|
"epoch": 8.297137664697864,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.756664698833492e-05,
|
|
"loss": 4.5274,
|
|
"mean_token_accuracy": 0.24185336530208587,
|
|
"num_tokens": 18917930.0,
|
|
"step": 9135
|
|
},
|
|
{
|
|
"entropy": 5.3359410762786865,
|
|
"epoch": 8.301681054066334,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 8.737135524624959e-05,
|
|
"loss": 4.5388,
|
|
"mean_token_accuracy": 0.24694875180721282,
|
|
"num_tokens": 18928506.0,
|
|
"step": 9140
|
|
},
|
|
{
|
|
"entropy": 5.290480470657348,
|
|
"epoch": 8.306224443434802,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 8.717652645919352e-05,
|
|
"loss": 4.4313,
|
|
"mean_token_accuracy": 0.2628057703375816,
|
|
"num_tokens": 18938811.0,
|
|
"step": 9145
|
|
},
|
|
{
|
|
"entropy": 5.230764150619507,
|
|
"epoch": 8.310767832803272,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 8.698216110788753e-05,
|
|
"loss": 4.4344,
|
|
"mean_token_accuracy": 0.25608912110328674,
|
|
"num_tokens": 18948487.0,
|
|
"step": 9150
|
|
},
|
|
{
|
|
"entropy": 5.269531345367431,
|
|
"epoch": 8.31531122217174,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.67882596719086e-05,
|
|
"loss": 4.4458,
|
|
"mean_token_accuracy": 0.2519275888800621,
|
|
"num_tokens": 18958021.0,
|
|
"step": 9155
|
|
},
|
|
{
|
|
"entropy": 5.258082342147827,
|
|
"epoch": 8.319854611540208,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 8.659482262968943e-05,
|
|
"loss": 4.4557,
|
|
"mean_token_accuracy": 0.25413386672735216,
|
|
"num_tokens": 18967937.0,
|
|
"step": 9160
|
|
},
|
|
{
|
|
"entropy": 5.301423788070679,
|
|
"epoch": 8.324398000908678,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 8.640185045851653e-05,
|
|
"loss": 4.4598,
|
|
"mean_token_accuracy": 0.2505499258637428,
|
|
"num_tokens": 18978119.0,
|
|
"step": 9165
|
|
},
|
|
{
|
|
"entropy": 5.246098327636719,
|
|
"epoch": 8.328941390277146,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 8.620934363452973e-05,
|
|
"loss": 4.4285,
|
|
"mean_token_accuracy": 0.25272568464279177,
|
|
"num_tokens": 18989036.0,
|
|
"step": 9170
|
|
},
|
|
{
|
|
"entropy": 5.301526832580566,
|
|
"epoch": 8.333484779645616,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.601730263272034e-05,
|
|
"loss": 4.5021,
|
|
"mean_token_accuracy": 0.2543795555830002,
|
|
"num_tokens": 18999561.0,
|
|
"step": 9175
|
|
},
|
|
{
|
|
"entropy": 5.227030992507935,
|
|
"epoch": 8.338028169014084,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 8.58257279269305e-05,
|
|
"loss": 4.3986,
|
|
"mean_token_accuracy": 0.2546402424573898,
|
|
"num_tokens": 19010086.0,
|
|
"step": 9180
|
|
},
|
|
{
|
|
"entropy": 5.207489252090454,
|
|
"epoch": 8.342571558382554,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 8.563461998985172e-05,
|
|
"loss": 4.354,
|
|
"mean_token_accuracy": 0.25839466154575347,
|
|
"num_tokens": 19019697.0,
|
|
"step": 9185
|
|
},
|
|
{
|
|
"entropy": 5.308208703994751,
|
|
"epoch": 8.347114947751022,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 8.544397929302389e-05,
|
|
"loss": 4.4846,
|
|
"mean_token_accuracy": 0.2531829908490181,
|
|
"num_tokens": 19030429.0,
|
|
"step": 9190
|
|
},
|
|
{
|
|
"entropy": 5.305572032928467,
|
|
"epoch": 8.351658337119492,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.525380630683389e-05,
|
|
"loss": 4.4677,
|
|
"mean_token_accuracy": 0.2506480202078819,
|
|
"num_tokens": 19041052.0,
|
|
"step": 9195
|
|
},
|
|
{
|
|
"entropy": 5.221908855438232,
|
|
"epoch": 8.35620172648796,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 8.506410150051478e-05,
|
|
"loss": 4.3985,
|
|
"mean_token_accuracy": 0.2621337309479713,
|
|
"num_tokens": 19050936.0,
|
|
"step": 9200
|
|
},
|
|
{
|
|
"entropy": 5.234786891937256,
|
|
"epoch": 8.360745115856428,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.487486534214423e-05,
|
|
"loss": 4.392,
|
|
"mean_token_accuracy": 0.2536278828978539,
|
|
"num_tokens": 19060884.0,
|
|
"step": 9205
|
|
},
|
|
{
|
|
"entropy": 5.255835390090942,
|
|
"epoch": 8.365288505224898,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 8.468609829864366e-05,
|
|
"loss": 4.4543,
|
|
"mean_token_accuracy": 0.24926721602678298,
|
|
"num_tokens": 19071405.0,
|
|
"step": 9210
|
|
},
|
|
{
|
|
"entropy": 5.3037773132324215,
|
|
"epoch": 8.369831894593366,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 8.449780083577697e-05,
|
|
"loss": 4.4244,
|
|
"mean_token_accuracy": 0.2524700775742531,
|
|
"num_tokens": 19081186.0,
|
|
"step": 9215
|
|
},
|
|
{
|
|
"entropy": 5.323787641525269,
|
|
"epoch": 8.374375283961836,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.430997341814947e-05,
|
|
"loss": 4.4815,
|
|
"mean_token_accuracy": 0.24872855246067047,
|
|
"num_tokens": 19091042.0,
|
|
"step": 9220
|
|
},
|
|
{
|
|
"entropy": 5.26722903251648,
|
|
"epoch": 8.378918673330304,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 8.41226165092066e-05,
|
|
"loss": 4.443,
|
|
"mean_token_accuracy": 0.24970578849315644,
|
|
"num_tokens": 19101408.0,
|
|
"step": 9225
|
|
},
|
|
{
|
|
"entropy": 5.2201636791229244,
|
|
"epoch": 8.383462062698774,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.393573057123299e-05,
|
|
"loss": 4.4333,
|
|
"mean_token_accuracy": 0.25462192445993426,
|
|
"num_tokens": 19111773.0,
|
|
"step": 9230
|
|
},
|
|
{
|
|
"entropy": 5.300150156021118,
|
|
"epoch": 8.388005452067242,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 8.374931606535099e-05,
|
|
"loss": 4.4697,
|
|
"mean_token_accuracy": 0.25380044281482694,
|
|
"num_tokens": 19121532.0,
|
|
"step": 9235
|
|
},
|
|
{
|
|
"entropy": 5.262150478363037,
|
|
"epoch": 8.392548841435712,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.356337345152002e-05,
|
|
"loss": 4.4915,
|
|
"mean_token_accuracy": 0.2442183256149292,
|
|
"num_tokens": 19131389.0,
|
|
"step": 9240
|
|
},
|
|
{
|
|
"entropy": 5.212237787246704,
|
|
"epoch": 8.39709223080418,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 8.33779031885348e-05,
|
|
"loss": 4.4259,
|
|
"mean_token_accuracy": 0.24955615550279617,
|
|
"num_tokens": 19141304.0,
|
|
"step": 9245
|
|
},
|
|
{
|
|
"entropy": 5.217935514450073,
|
|
"epoch": 8.401635620172648,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 8.319290573402494e-05,
|
|
"loss": 4.4056,
|
|
"mean_token_accuracy": 0.24827489107847214,
|
|
"num_tokens": 19151039.0,
|
|
"step": 9250
|
|
},
|
|
{
|
|
"entropy": 5.232212066650391,
|
|
"epoch": 8.406179009541118,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.300838154445314e-05,
|
|
"loss": 4.4095,
|
|
"mean_token_accuracy": 0.25995958894491195,
|
|
"num_tokens": 19160909.0,
|
|
"step": 9255
|
|
},
|
|
{
|
|
"entropy": 5.291892385482788,
|
|
"epoch": 8.410722398909586,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 8.282433107511466e-05,
|
|
"loss": 4.4322,
|
|
"mean_token_accuracy": 0.24841741919517518,
|
|
"num_tokens": 19171906.0,
|
|
"step": 9260
|
|
},
|
|
{
|
|
"entropy": 5.2931029319763185,
|
|
"epoch": 8.415265788278056,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.264075478013552e-05,
|
|
"loss": 4.4814,
|
|
"mean_token_accuracy": 0.25569123327732085,
|
|
"num_tokens": 19182688.0,
|
|
"step": 9265
|
|
},
|
|
{
|
|
"entropy": 5.267188644409179,
|
|
"epoch": 8.419809177646524,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.245765311247216e-05,
|
|
"loss": 4.3931,
|
|
"mean_token_accuracy": 0.25487874150276185,
|
|
"num_tokens": 19192429.0,
|
|
"step": 9270
|
|
},
|
|
{
|
|
"entropy": 5.3033746719360355,
|
|
"epoch": 8.424352567014994,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 8.227502652390968e-05,
|
|
"loss": 4.531,
|
|
"mean_token_accuracy": 0.2442270889878273,
|
|
"num_tokens": 19202454.0,
|
|
"step": 9275
|
|
},
|
|
{
|
|
"entropy": 5.341376161575317,
|
|
"epoch": 8.428895956383462,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.2092875465061e-05,
|
|
"loss": 4.5286,
|
|
"mean_token_accuracy": 0.2468560203909874,
|
|
"num_tokens": 19213057.0,
|
|
"step": 9280
|
|
},
|
|
{
|
|
"entropy": 5.204781770706177,
|
|
"epoch": 8.43343934575193,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.191120038536575e-05,
|
|
"loss": 4.3561,
|
|
"mean_token_accuracy": 0.2595460280776024,
|
|
"num_tokens": 19222417.0,
|
|
"step": 9285
|
|
},
|
|
{
|
|
"entropy": 5.289634847640992,
|
|
"epoch": 8.4379827351204,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.173000173308921e-05,
|
|
"loss": 4.4796,
|
|
"mean_token_accuracy": 0.24240662008523942,
|
|
"num_tokens": 19233743.0,
|
|
"step": 9290
|
|
},
|
|
{
|
|
"entropy": 5.331985521316528,
|
|
"epoch": 8.442526124488868,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 8.154927995532098e-05,
|
|
"loss": 4.6011,
|
|
"mean_token_accuracy": 0.24064554870128632,
|
|
"num_tokens": 19245000.0,
|
|
"step": 9295
|
|
},
|
|
{
|
|
"entropy": 5.341682100296021,
|
|
"epoch": 8.447069513857338,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 8.136903549797408e-05,
|
|
"loss": 4.4606,
|
|
"mean_token_accuracy": 0.2512967213988304,
|
|
"num_tokens": 19255738.0,
|
|
"step": 9300
|
|
},
|
|
{
|
|
"entropy": 5.3091188907623295,
|
|
"epoch": 8.451612903225806,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.118926880578378e-05,
|
|
"loss": 4.5063,
|
|
"mean_token_accuracy": 0.2552332878112793,
|
|
"num_tokens": 19266268.0,
|
|
"step": 9305
|
|
},
|
|
{
|
|
"entropy": 5.404523086547852,
|
|
"epoch": 8.456156292594276,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.100998032230656e-05,
|
|
"loss": 4.5429,
|
|
"mean_token_accuracy": 0.24825963675975798,
|
|
"num_tokens": 19276721.0,
|
|
"step": 9310
|
|
},
|
|
{
|
|
"entropy": 5.280344724655151,
|
|
"epoch": 8.460699681962744,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.083117048991895e-05,
|
|
"loss": 4.4212,
|
|
"mean_token_accuracy": 0.25098801106214524,
|
|
"num_tokens": 19287241.0,
|
|
"step": 9315
|
|
},
|
|
{
|
|
"entropy": 5.275425004959106,
|
|
"epoch": 8.465243071331214,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 8.065283974981638e-05,
|
|
"loss": 4.5367,
|
|
"mean_token_accuracy": 0.24539594799280168,
|
|
"num_tokens": 19298065.0,
|
|
"step": 9320
|
|
},
|
|
{
|
|
"entropy": 5.313815927505493,
|
|
"epoch": 8.469786460699682,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 8.047498854201222e-05,
|
|
"loss": 4.5135,
|
|
"mean_token_accuracy": 0.2389595165848732,
|
|
"num_tokens": 19308108.0,
|
|
"step": 9325
|
|
},
|
|
{
|
|
"entropy": 5.268885660171509,
|
|
"epoch": 8.47432985006815,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 8.029761730533678e-05,
|
|
"loss": 4.4851,
|
|
"mean_token_accuracy": 0.24774858504533767,
|
|
"num_tokens": 19317909.0,
|
|
"step": 9330
|
|
},
|
|
{
|
|
"entropy": 5.310593843460083,
|
|
"epoch": 8.47887323943662,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 8.012072647743569e-05,
|
|
"loss": 4.5273,
|
|
"mean_token_accuracy": 0.2554711028933525,
|
|
"num_tokens": 19328278.0,
|
|
"step": 9335
|
|
},
|
|
{
|
|
"entropy": 5.210761070251465,
|
|
"epoch": 8.483416628805088,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.994431649476974e-05,
|
|
"loss": 4.4267,
|
|
"mean_token_accuracy": 0.25453328639268874,
|
|
"num_tokens": 19338995.0,
|
|
"step": 9340
|
|
},
|
|
{
|
|
"entropy": 5.2926231861114506,
|
|
"epoch": 8.487960018173558,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.976838779261285e-05,
|
|
"loss": 4.469,
|
|
"mean_token_accuracy": 0.2510872110724449,
|
|
"num_tokens": 19349177.0,
|
|
"step": 9345
|
|
},
|
|
{
|
|
"entropy": 5.388121509552002,
|
|
"epoch": 8.492503407542026,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.959294080505179e-05,
|
|
"loss": 4.5063,
|
|
"mean_token_accuracy": 0.2512108087539673,
|
|
"num_tokens": 19358871.0,
|
|
"step": 9350
|
|
},
|
|
{
|
|
"entropy": 5.292498970031739,
|
|
"epoch": 8.497046796910496,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.941797596498435e-05,
|
|
"loss": 4.5179,
|
|
"mean_token_accuracy": 0.2512828722596169,
|
|
"num_tokens": 19368594.0,
|
|
"step": 9355
|
|
},
|
|
{
|
|
"entropy": 5.244577789306641,
|
|
"epoch": 8.501590186278964,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 7.924349370411904e-05,
|
|
"loss": 4.4097,
|
|
"mean_token_accuracy": 0.253965300321579,
|
|
"num_tokens": 19379973.0,
|
|
"step": 9360
|
|
},
|
|
{
|
|
"entropy": 5.288029050827026,
|
|
"epoch": 8.506133575647432,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.906949445297342e-05,
|
|
"loss": 4.4324,
|
|
"mean_token_accuracy": 0.2548566982150078,
|
|
"num_tokens": 19390762.0,
|
|
"step": 9365
|
|
},
|
|
{
|
|
"entropy": 5.2856827735900875,
|
|
"epoch": 8.510676965015902,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.889597864087336e-05,
|
|
"loss": 4.4997,
|
|
"mean_token_accuracy": 0.2542933315038681,
|
|
"num_tokens": 19401918.0,
|
|
"step": 9370
|
|
},
|
|
{
|
|
"entropy": 5.233430290222168,
|
|
"epoch": 8.51522035438437,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 7.872294669595185e-05,
|
|
"loss": 4.425,
|
|
"mean_token_accuracy": 0.25818985849618914,
|
|
"num_tokens": 19411973.0,
|
|
"step": 9375
|
|
},
|
|
{
|
|
"entropy": 5.283972454071045,
|
|
"epoch": 8.51976374375284,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.855039904514808e-05,
|
|
"loss": 4.4254,
|
|
"mean_token_accuracy": 0.249590140581131,
|
|
"num_tokens": 19422888.0,
|
|
"step": 9380
|
|
},
|
|
{
|
|
"entropy": 5.351130723953247,
|
|
"epoch": 8.524307133121308,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.837833611420618e-05,
|
|
"loss": 4.5004,
|
|
"mean_token_accuracy": 0.252889609336853,
|
|
"num_tokens": 19433611.0,
|
|
"step": 9385
|
|
},
|
|
{
|
|
"entropy": 5.195641565322876,
|
|
"epoch": 8.528850522489778,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.820675832767435e-05,
|
|
"loss": 4.4141,
|
|
"mean_token_accuracy": 0.2579168826341629,
|
|
"num_tokens": 19443712.0,
|
|
"step": 9390
|
|
},
|
|
{
|
|
"entropy": 5.2451292991638185,
|
|
"epoch": 8.533393911858246,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 7.803566610890365e-05,
|
|
"loss": 4.3581,
|
|
"mean_token_accuracy": 0.25387939512729646,
|
|
"num_tokens": 19453410.0,
|
|
"step": 9395
|
|
},
|
|
{
|
|
"entropy": 5.237646818161011,
|
|
"epoch": 8.537937301226716,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 7.786505988004722e-05,
|
|
"loss": 4.4077,
|
|
"mean_token_accuracy": 0.2623248934745789,
|
|
"num_tokens": 19463306.0,
|
|
"step": 9400
|
|
},
|
|
{
|
|
"entropy": 5.305750370025635,
|
|
"epoch": 8.542480690595184,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.769494006205894e-05,
|
|
"loss": 4.5142,
|
|
"mean_token_accuracy": 0.2545729920268059,
|
|
"num_tokens": 19474127.0,
|
|
"step": 9405
|
|
},
|
|
{
|
|
"entropy": 5.3486841201782225,
|
|
"epoch": 8.547024079963652,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 7.752530707469252e-05,
|
|
"loss": 4.5402,
|
|
"mean_token_accuracy": 0.24994128346443176,
|
|
"num_tokens": 19486472.0,
|
|
"step": 9410
|
|
},
|
|
{
|
|
"entropy": 5.277399969100952,
|
|
"epoch": 8.551567469332122,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.73561613365005e-05,
|
|
"loss": 4.4718,
|
|
"mean_token_accuracy": 0.25798999667167666,
|
|
"num_tokens": 19497130.0,
|
|
"step": 9415
|
|
},
|
|
{
|
|
"entropy": 5.241492557525635,
|
|
"epoch": 8.55611085870059,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 7.718750326483321e-05,
|
|
"loss": 4.3874,
|
|
"mean_token_accuracy": 0.25902165472507477,
|
|
"num_tokens": 19507321.0,
|
|
"step": 9420
|
|
},
|
|
{
|
|
"entropy": 5.3227612495422365,
|
|
"epoch": 8.56065424806906,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 7.701933327583766e-05,
|
|
"loss": 4.5146,
|
|
"mean_token_accuracy": 0.2466563895344734,
|
|
"num_tokens": 19516244.0,
|
|
"step": 9425
|
|
},
|
|
{
|
|
"entropy": 5.254840278625489,
|
|
"epoch": 8.565197637437528,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.685165178445658e-05,
|
|
"loss": 4.415,
|
|
"mean_token_accuracy": 0.2618802338838577,
|
|
"num_tokens": 19526594.0,
|
|
"step": 9430
|
|
},
|
|
{
|
|
"entropy": 5.23323802947998,
|
|
"epoch": 8.569741026805998,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.668445920442732e-05,
|
|
"loss": 4.4269,
|
|
"mean_token_accuracy": 0.24409262835979462,
|
|
"num_tokens": 19536835.0,
|
|
"step": 9435
|
|
},
|
|
{
|
|
"entropy": 5.243249416351318,
|
|
"epoch": 8.574284416174466,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.651775594828112e-05,
|
|
"loss": 4.516,
|
|
"mean_token_accuracy": 0.2444480836391449,
|
|
"num_tokens": 19546474.0,
|
|
"step": 9440
|
|
},
|
|
{
|
|
"entropy": 5.348822975158692,
|
|
"epoch": 8.578827805542936,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.635154242734146e-05,
|
|
"loss": 4.5286,
|
|
"mean_token_accuracy": 0.2427505671977997,
|
|
"num_tokens": 19556604.0,
|
|
"step": 9445
|
|
},
|
|
{
|
|
"entropy": 5.2132128238677975,
|
|
"epoch": 8.583371194911404,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 7.618581905172385e-05,
|
|
"loss": 4.4095,
|
|
"mean_token_accuracy": 0.25719524174928665,
|
|
"num_tokens": 19566824.0,
|
|
"step": 9450
|
|
},
|
|
{
|
|
"entropy": 5.25755763053894,
|
|
"epoch": 8.587914584279872,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.60205862303342e-05,
|
|
"loss": 4.4047,
|
|
"mean_token_accuracy": 0.2573779091238976,
|
|
"num_tokens": 19576917.0,
|
|
"step": 9455
|
|
},
|
|
{
|
|
"entropy": 5.279919528961182,
|
|
"epoch": 8.592457973648342,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 7.585584437086806e-05,
|
|
"loss": 4.4764,
|
|
"mean_token_accuracy": 0.24193051755428313,
|
|
"num_tokens": 19587651.0,
|
|
"step": 9460
|
|
},
|
|
{
|
|
"entropy": 5.257141017913819,
|
|
"epoch": 8.59700136301681,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 7.569159387980958e-05,
|
|
"loss": 4.4495,
|
|
"mean_token_accuracy": 0.25582503229379655,
|
|
"num_tokens": 19598421.0,
|
|
"step": 9465
|
|
},
|
|
{
|
|
"entropy": 5.218360948562622,
|
|
"epoch": 8.60154475238528,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.552783516243056e-05,
|
|
"loss": 4.4068,
|
|
"mean_token_accuracy": 0.26091408878564837,
|
|
"num_tokens": 19608715.0,
|
|
"step": 9470
|
|
},
|
|
{
|
|
"entropy": 5.284559679031372,
|
|
"epoch": 8.606088141753748,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.536456862278936e-05,
|
|
"loss": 4.4717,
|
|
"mean_token_accuracy": 0.2459798961877823,
|
|
"num_tokens": 19619676.0,
|
|
"step": 9475
|
|
},
|
|
{
|
|
"entropy": 5.290702676773071,
|
|
"epoch": 8.610631531122218,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 7.520179466372988e-05,
|
|
"loss": 4.4847,
|
|
"mean_token_accuracy": 0.2562276303768158,
|
|
"num_tokens": 19630837.0,
|
|
"step": 9480
|
|
},
|
|
{
|
|
"entropy": 5.216637372970581,
|
|
"epoch": 8.615174920490686,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 7.503951368688068e-05,
|
|
"loss": 4.3495,
|
|
"mean_token_accuracy": 0.2588009059429169,
|
|
"num_tokens": 19640607.0,
|
|
"step": 9485
|
|
},
|
|
{
|
|
"entropy": 5.293661165237427,
|
|
"epoch": 8.619718309859154,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.487772609265401e-05,
|
|
"loss": 4.5338,
|
|
"mean_token_accuracy": 0.24394749701023102,
|
|
"num_tokens": 19651354.0,
|
|
"step": 9490
|
|
},
|
|
{
|
|
"entropy": 5.242184448242187,
|
|
"epoch": 8.624261699227624,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 7.471643228024466e-05,
|
|
"loss": 4.4053,
|
|
"mean_token_accuracy": 0.25868902206420896,
|
|
"num_tokens": 19661266.0,
|
|
"step": 9495
|
|
},
|
|
{
|
|
"entropy": 5.266915225982666,
|
|
"epoch": 8.628805088596092,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.455563264762903e-05,
|
|
"loss": 4.4688,
|
|
"mean_token_accuracy": 0.2532314345240593,
|
|
"num_tokens": 19671275.0,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"epoch": 8.628805088596092,
|
|
"eval_entropy": 5.1686581844507264,
|
|
"eval_loss": 5.676290512084961,
|
|
"eval_mean_token_accuracy": 0.17483005605118218,
|
|
"eval_num_tokens": 19671275.0,
|
|
"eval_runtime": 2.2049,
|
|
"eval_samples_per_second": 1557.874,
|
|
"eval_steps_per_second": 195.018,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"entropy": 5.259424448013306,
|
|
"epoch": 8.633348477964562,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 7.439532759156422e-05,
|
|
"loss": 4.3399,
|
|
"mean_token_accuracy": 0.26629735678434374,
|
|
"num_tokens": 19682360.0,
|
|
"step": 9505
|
|
},
|
|
{
|
|
"entropy": 5.302112197875976,
|
|
"epoch": 8.63789186733303,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.423551750758711e-05,
|
|
"loss": 4.489,
|
|
"mean_token_accuracy": 0.24617168456315994,
|
|
"num_tokens": 19692393.0,
|
|
"step": 9510
|
|
},
|
|
{
|
|
"entropy": 5.2860442161560055,
|
|
"epoch": 8.6424352567015,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.407620279001315e-05,
|
|
"loss": 4.4044,
|
|
"mean_token_accuracy": 0.2555647730827332,
|
|
"num_tokens": 19702045.0,
|
|
"step": 9515
|
|
},
|
|
{
|
|
"entropy": 5.32166600227356,
|
|
"epoch": 8.646978646069968,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 7.391738383193553e-05,
|
|
"loss": 4.5368,
|
|
"mean_token_accuracy": 0.24284496903419495,
|
|
"num_tokens": 19713038.0,
|
|
"step": 9520
|
|
},
|
|
{
|
|
"entropy": 5.218878936767578,
|
|
"epoch": 8.651522035438438,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 7.375906102522423e-05,
|
|
"loss": 4.4609,
|
|
"mean_token_accuracy": 0.2536552160978317,
|
|
"num_tokens": 19723708.0,
|
|
"step": 9525
|
|
},
|
|
{
|
|
"entropy": 5.279366064071655,
|
|
"epoch": 8.656065424806906,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 7.360123476052519e-05,
|
|
"loss": 4.5271,
|
|
"mean_token_accuracy": 0.24775732904672623,
|
|
"num_tokens": 19735271.0,
|
|
"step": 9530
|
|
},
|
|
{
|
|
"entropy": 5.321377611160278,
|
|
"epoch": 8.660608814175374,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.344390542725882e-05,
|
|
"loss": 4.4853,
|
|
"mean_token_accuracy": 0.24715246856212617,
|
|
"num_tokens": 19744593.0,
|
|
"step": 9535
|
|
},
|
|
{
|
|
"entropy": 5.203895521163941,
|
|
"epoch": 8.665152203543844,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 7.328707341361975e-05,
|
|
"loss": 4.394,
|
|
"mean_token_accuracy": 0.26336713284254076,
|
|
"num_tokens": 19754600.0,
|
|
"step": 9540
|
|
},
|
|
{
|
|
"entropy": 5.29565920829773,
|
|
"epoch": 8.669695592912312,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.313073910657535e-05,
|
|
"loss": 4.5113,
|
|
"mean_token_accuracy": 0.2511609077453613,
|
|
"num_tokens": 19766376.0,
|
|
"step": 9545
|
|
},
|
|
{
|
|
"entropy": 5.244127082824707,
|
|
"epoch": 8.674238982280782,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.297490289186495e-05,
|
|
"loss": 4.3843,
|
|
"mean_token_accuracy": 0.25854971408843996,
|
|
"num_tokens": 19776015.0,
|
|
"step": 9550
|
|
},
|
|
{
|
|
"entropy": 5.254857301712036,
|
|
"epoch": 8.67878237164925,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.281956515399891e-05,
|
|
"loss": 4.4053,
|
|
"mean_token_accuracy": 0.25365958362817764,
|
|
"num_tokens": 19785794.0,
|
|
"step": 9555
|
|
},
|
|
{
|
|
"entropy": 5.296278476715088,
|
|
"epoch": 8.68332576101772,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 7.266472627625776e-05,
|
|
"loss": 4.4391,
|
|
"mean_token_accuracy": 0.25210391283035277,
|
|
"num_tokens": 19795202.0,
|
|
"step": 9560
|
|
},
|
|
{
|
|
"entropy": 5.28316068649292,
|
|
"epoch": 8.687869150386188,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 7.251038664069094e-05,
|
|
"loss": 4.46,
|
|
"mean_token_accuracy": 0.24523129612207412,
|
|
"num_tokens": 19805623.0,
|
|
"step": 9565
|
|
},
|
|
{
|
|
"entropy": 5.287612867355347,
|
|
"epoch": 8.692412539754656,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.235654662811618e-05,
|
|
"loss": 4.4373,
|
|
"mean_token_accuracy": 0.2486402213573456,
|
|
"num_tokens": 19815792.0,
|
|
"step": 9570
|
|
},
|
|
{
|
|
"entropy": 5.252313613891602,
|
|
"epoch": 8.696955929123126,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.22032066181184e-05,
|
|
"loss": 4.4213,
|
|
"mean_token_accuracy": 0.2512613371014595,
|
|
"num_tokens": 19825460.0,
|
|
"step": 9575
|
|
},
|
|
{
|
|
"entropy": 5.2559041500091555,
|
|
"epoch": 8.701499318491594,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.205036698904888e-05,
|
|
"loss": 4.4931,
|
|
"mean_token_accuracy": 0.24949656128883363,
|
|
"num_tokens": 19836630.0,
|
|
"step": 9580
|
|
},
|
|
{
|
|
"entropy": 5.268910837173462,
|
|
"epoch": 8.706042707860064,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 7.18980281180242e-05,
|
|
"loss": 4.4345,
|
|
"mean_token_accuracy": 0.2552192881703377,
|
|
"num_tokens": 19846445.0,
|
|
"step": 9585
|
|
},
|
|
{
|
|
"entropy": 5.246001434326172,
|
|
"epoch": 8.710586097228532,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.174619038092537e-05,
|
|
"loss": 4.4254,
|
|
"mean_token_accuracy": 0.25642859041690824,
|
|
"num_tokens": 19856738.0,
|
|
"step": 9590
|
|
},
|
|
{
|
|
"entropy": 5.246344995498657,
|
|
"epoch": 8.715129486597002,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 7.159485415239685e-05,
|
|
"loss": 4.4705,
|
|
"mean_token_accuracy": 0.24833889901638032,
|
|
"num_tokens": 19866134.0,
|
|
"step": 9595
|
|
},
|
|
{
|
|
"entropy": 5.236944580078125,
|
|
"epoch": 8.71967287596547,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.144401980584584e-05,
|
|
"loss": 4.3848,
|
|
"mean_token_accuracy": 0.2575332194566727,
|
|
"num_tokens": 19877384.0,
|
|
"step": 9600
|
|
},
|
|
{
|
|
"entropy": 5.293432569503784,
|
|
"epoch": 8.72421626533394,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 7.129368771344108e-05,
|
|
"loss": 4.4164,
|
|
"mean_token_accuracy": 0.2562857583165169,
|
|
"num_tokens": 19887836.0,
|
|
"step": 9605
|
|
},
|
|
{
|
|
"entropy": 5.273995113372803,
|
|
"epoch": 8.728759654702408,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 7.114385824611203e-05,
|
|
"loss": 4.4516,
|
|
"mean_token_accuracy": 0.24891214221715927,
|
|
"num_tokens": 19897323.0,
|
|
"step": 9610
|
|
},
|
|
{
|
|
"entropy": 5.256911706924439,
|
|
"epoch": 8.733303044070876,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 7.099453177354797e-05,
|
|
"loss": 4.4825,
|
|
"mean_token_accuracy": 0.24081410020589827,
|
|
"num_tokens": 19908959.0,
|
|
"step": 9615
|
|
},
|
|
{
|
|
"entropy": 5.254597806930542,
|
|
"epoch": 8.737846433439346,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.084570866419727e-05,
|
|
"loss": 4.389,
|
|
"mean_token_accuracy": 0.26243137419223783,
|
|
"num_tokens": 19918918.0,
|
|
"step": 9620
|
|
},
|
|
{
|
|
"entropy": 5.270342254638672,
|
|
"epoch": 8.742389822807814,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 7.069738928526597e-05,
|
|
"loss": 4.5125,
|
|
"mean_token_accuracy": 0.25395732223987577,
|
|
"num_tokens": 19928804.0,
|
|
"step": 9625
|
|
},
|
|
{
|
|
"entropy": 5.278969192504883,
|
|
"epoch": 8.746933212176284,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.054957400271751e-05,
|
|
"loss": 4.4564,
|
|
"mean_token_accuracy": 0.2508986577391624,
|
|
"num_tokens": 19938706.0,
|
|
"step": 9630
|
|
},
|
|
{
|
|
"entropy": 5.215254783630371,
|
|
"epoch": 8.751476601544752,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 7.040226318127142e-05,
|
|
"loss": 4.3552,
|
|
"mean_token_accuracy": 0.2667869046330452,
|
|
"num_tokens": 19948548.0,
|
|
"step": 9635
|
|
},
|
|
{
|
|
"entropy": 5.212531232833863,
|
|
"epoch": 8.756019990913222,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 7.025545718440245e-05,
|
|
"loss": 4.4284,
|
|
"mean_token_accuracy": 0.2537317410111427,
|
|
"num_tokens": 19958568.0,
|
|
"step": 9640
|
|
},
|
|
{
|
|
"entropy": 5.2738086700439455,
|
|
"epoch": 8.76056338028169,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.01091563743398e-05,
|
|
"loss": 4.4573,
|
|
"mean_token_accuracy": 0.25323972553014756,
|
|
"num_tokens": 19968889.0,
|
|
"step": 9645
|
|
},
|
|
{
|
|
"entropy": 5.2645806789398195,
|
|
"epoch": 8.76510676965016,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.996336111206619e-05,
|
|
"loss": 4.4503,
|
|
"mean_token_accuracy": 0.245570570230484,
|
|
"num_tokens": 19978813.0,
|
|
"step": 9650
|
|
},
|
|
{
|
|
"entropy": 5.241973543167115,
|
|
"epoch": 8.769650159018628,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.981807175731698e-05,
|
|
"loss": 4.3977,
|
|
"mean_token_accuracy": 0.2553370535373688,
|
|
"num_tokens": 19988689.0,
|
|
"step": 9655
|
|
},
|
|
{
|
|
"entropy": 5.278346729278565,
|
|
"epoch": 8.774193548387096,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 6.967328866857915e-05,
|
|
"loss": 4.4431,
|
|
"mean_token_accuracy": 0.25309514701366426,
|
|
"num_tokens": 19999006.0,
|
|
"step": 9660
|
|
},
|
|
{
|
|
"entropy": 5.304649925231933,
|
|
"epoch": 8.778736937755566,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 6.952901220309063e-05,
|
|
"loss": 4.4931,
|
|
"mean_token_accuracy": 0.2471567362546921,
|
|
"num_tokens": 20009038.0,
|
|
"step": 9665
|
|
},
|
|
{
|
|
"entropy": 5.321915960311889,
|
|
"epoch": 8.783280327124034,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 6.938524271683918e-05,
|
|
"loss": 4.5567,
|
|
"mean_token_accuracy": 0.23862184286117555,
|
|
"num_tokens": 20020908.0,
|
|
"step": 9670
|
|
},
|
|
{
|
|
"entropy": 5.339586496353149,
|
|
"epoch": 8.787823716492504,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 6.924198056456185e-05,
|
|
"loss": 4.4852,
|
|
"mean_token_accuracy": 0.25163899809122087,
|
|
"num_tokens": 20032441.0,
|
|
"step": 9675
|
|
},
|
|
{
|
|
"entropy": 5.320588302612305,
|
|
"epoch": 8.792367105860972,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.90992260997436e-05,
|
|
"loss": 4.5523,
|
|
"mean_token_accuracy": 0.24040723890066146,
|
|
"num_tokens": 20043772.0,
|
|
"step": 9680
|
|
},
|
|
{
|
|
"entropy": 5.232398557662964,
|
|
"epoch": 8.796910495229442,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 6.895697967461699e-05,
|
|
"loss": 4.4358,
|
|
"mean_token_accuracy": 0.25236757695674894,
|
|
"num_tokens": 20053058.0,
|
|
"step": 9685
|
|
},
|
|
{
|
|
"entropy": 5.272405624389648,
|
|
"epoch": 8.80145388459791,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.881524164016087e-05,
|
|
"loss": 4.5217,
|
|
"mean_token_accuracy": 0.24486826211214066,
|
|
"num_tokens": 20064289.0,
|
|
"step": 9690
|
|
},
|
|
{
|
|
"entropy": 5.283498477935791,
|
|
"epoch": 8.805997273966378,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 6.867401234609985e-05,
|
|
"loss": 4.4443,
|
|
"mean_token_accuracy": 0.2581868276000023,
|
|
"num_tokens": 20074813.0,
|
|
"step": 9695
|
|
},
|
|
{
|
|
"entropy": 5.29147162437439,
|
|
"epoch": 8.810540663334848,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.853329214090304e-05,
|
|
"loss": 4.4483,
|
|
"mean_token_accuracy": 0.24727970063686372,
|
|
"num_tokens": 20085019.0,
|
|
"step": 9700
|
|
},
|
|
{
|
|
"entropy": 5.330164909362793,
|
|
"epoch": 8.815084052703316,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.839308137178363e-05,
|
|
"loss": 4.4835,
|
|
"mean_token_accuracy": 0.2568948701024055,
|
|
"num_tokens": 20094773.0,
|
|
"step": 9705
|
|
},
|
|
{
|
|
"entropy": 5.27206072807312,
|
|
"epoch": 8.819627442071786,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.825338038469778e-05,
|
|
"loss": 4.4181,
|
|
"mean_token_accuracy": 0.25122364014387133,
|
|
"num_tokens": 20105612.0,
|
|
"step": 9710
|
|
},
|
|
{
|
|
"entropy": 5.247729158401489,
|
|
"epoch": 8.824170831440254,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 6.811418952434375e-05,
|
|
"loss": 4.4558,
|
|
"mean_token_accuracy": 0.24918756037950515,
|
|
"num_tokens": 20117032.0,
|
|
"step": 9715
|
|
},
|
|
{
|
|
"entropy": 5.315434789657592,
|
|
"epoch": 8.828714220808724,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 6.797550913416111e-05,
|
|
"loss": 4.4437,
|
|
"mean_token_accuracy": 0.25846472531557085,
|
|
"num_tokens": 20127539.0,
|
|
"step": 9720
|
|
},
|
|
{
|
|
"entropy": 5.278774976730347,
|
|
"epoch": 8.833257610177192,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.783733955633004e-05,
|
|
"loss": 4.467,
|
|
"mean_token_accuracy": 0.2511288031935692,
|
|
"num_tokens": 20138430.0,
|
|
"step": 9725
|
|
},
|
|
{
|
|
"entropy": 5.235414361953735,
|
|
"epoch": 8.837800999545662,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 6.769968113177016e-05,
|
|
"loss": 4.4386,
|
|
"mean_token_accuracy": 0.24691525250673294,
|
|
"num_tokens": 20149261.0,
|
|
"step": 9730
|
|
},
|
|
{
|
|
"entropy": 5.283322477340699,
|
|
"epoch": 8.84234438891413,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 6.756253420013996e-05,
|
|
"loss": 4.4057,
|
|
"mean_token_accuracy": 0.2616446644067764,
|
|
"num_tokens": 20158617.0,
|
|
"step": 9735
|
|
},
|
|
{
|
|
"entropy": 5.258287048339843,
|
|
"epoch": 8.846887778282598,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.742589909983586e-05,
|
|
"loss": 4.456,
|
|
"mean_token_accuracy": 0.25269039273262023,
|
|
"num_tokens": 20169028.0,
|
|
"step": 9740
|
|
},
|
|
{
|
|
"entropy": 5.264219188690186,
|
|
"epoch": 8.851431167651068,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 6.728977616799144e-05,
|
|
"loss": 4.3998,
|
|
"mean_token_accuracy": 0.2650249317288399,
|
|
"num_tokens": 20178926.0,
|
|
"step": 9745
|
|
},
|
|
{
|
|
"entropy": 5.267004489898682,
|
|
"epoch": 8.855974557019536,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.715416574047645e-05,
|
|
"loss": 4.374,
|
|
"mean_token_accuracy": 0.26413504332304,
|
|
"num_tokens": 20188602.0,
|
|
"step": 9750
|
|
},
|
|
{
|
|
"entropy": 5.230362892150879,
|
|
"epoch": 8.860517946388006,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.701906815189613e-05,
|
|
"loss": 4.4229,
|
|
"mean_token_accuracy": 0.2513081178069115,
|
|
"num_tokens": 20198623.0,
|
|
"step": 9755
|
|
},
|
|
{
|
|
"entropy": 5.294963598251343,
|
|
"epoch": 8.865061335756474,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 6.688448373559037e-05,
|
|
"loss": 4.46,
|
|
"mean_token_accuracy": 0.25051717311143873,
|
|
"num_tokens": 20208627.0,
|
|
"step": 9760
|
|
},
|
|
{
|
|
"entropy": 5.3172932147979735,
|
|
"epoch": 8.869604725124944,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.675041282363286e-05,
|
|
"loss": 4.4933,
|
|
"mean_token_accuracy": 0.24814092963933945,
|
|
"num_tokens": 20218920.0,
|
|
"step": 9765
|
|
},
|
|
{
|
|
"entropy": 5.271107912063599,
|
|
"epoch": 8.874148114493412,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.661685574683021e-05,
|
|
"loss": 4.4722,
|
|
"mean_token_accuracy": 0.25141964852809906,
|
|
"num_tokens": 20229938.0,
|
|
"step": 9770
|
|
},
|
|
{
|
|
"entropy": 5.2690184116363525,
|
|
"epoch": 8.87869150386188,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.648381283472128e-05,
|
|
"loss": 4.475,
|
|
"mean_token_accuracy": 0.24809191673994063,
|
|
"num_tokens": 20240135.0,
|
|
"step": 9775
|
|
},
|
|
{
|
|
"entropy": 5.210130786895752,
|
|
"epoch": 8.88323489323035,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 6.635128441557615e-05,
|
|
"loss": 4.4503,
|
|
"mean_token_accuracy": 0.26131738126277926,
|
|
"num_tokens": 20250239.0,
|
|
"step": 9780
|
|
},
|
|
{
|
|
"entropy": 5.276074981689453,
|
|
"epoch": 8.887778282598818,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.621927081639563e-05,
|
|
"loss": 4.5101,
|
|
"mean_token_accuracy": 0.2506806939840317,
|
|
"num_tokens": 20260498.0,
|
|
"step": 9785
|
|
},
|
|
{
|
|
"entropy": 5.279958868026734,
|
|
"epoch": 8.892321671967288,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.608777236291003e-05,
|
|
"loss": 4.5216,
|
|
"mean_token_accuracy": 0.24684226214885713,
|
|
"num_tokens": 20271830.0,
|
|
"step": 9790
|
|
},
|
|
{
|
|
"entropy": 5.274948692321777,
|
|
"epoch": 8.896865061335756,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 6.59567893795788e-05,
|
|
"loss": 4.494,
|
|
"mean_token_accuracy": 0.2508978024125099,
|
|
"num_tokens": 20282519.0,
|
|
"step": 9795
|
|
},
|
|
{
|
|
"entropy": 5.226023435592651,
|
|
"epoch": 8.901408450704226,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.58263221895894e-05,
|
|
"loss": 4.4202,
|
|
"mean_token_accuracy": 0.25162608325481417,
|
|
"num_tokens": 20292285.0,
|
|
"step": 9800
|
|
},
|
|
{
|
|
"entropy": 5.3129981517791744,
|
|
"epoch": 8.905951840072694,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 6.569637111485666e-05,
|
|
"loss": 4.471,
|
|
"mean_token_accuracy": 0.25291620939970016,
|
|
"num_tokens": 20302191.0,
|
|
"step": 9805
|
|
},
|
|
{
|
|
"entropy": 5.28643012046814,
|
|
"epoch": 8.910495229441164,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 6.556693647602191e-05,
|
|
"loss": 4.4333,
|
|
"mean_token_accuracy": 0.25766642987728117,
|
|
"num_tokens": 20313839.0,
|
|
"step": 9810
|
|
},
|
|
{
|
|
"entropy": 5.229101705551147,
|
|
"epoch": 8.915038618809632,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 6.543801859245229e-05,
|
|
"loss": 4.4784,
|
|
"mean_token_accuracy": 0.2518684729933739,
|
|
"num_tokens": 20324016.0,
|
|
"step": 9815
|
|
},
|
|
{
|
|
"entropy": 5.279875564575195,
|
|
"epoch": 8.9195820081781,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.530961778223988e-05,
|
|
"loss": 4.4332,
|
|
"mean_token_accuracy": 0.2516744315624237,
|
|
"num_tokens": 20334085.0,
|
|
"step": 9820
|
|
},
|
|
{
|
|
"entropy": 5.312024354934692,
|
|
"epoch": 8.92412539754657,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.518173436220089e-05,
|
|
"loss": 4.5042,
|
|
"mean_token_accuracy": 0.24921342581510544,
|
|
"num_tokens": 20344647.0,
|
|
"step": 9825
|
|
},
|
|
{
|
|
"entropy": 5.324291515350342,
|
|
"epoch": 8.928668786915038,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.505436864787489e-05,
|
|
"loss": 4.4792,
|
|
"mean_token_accuracy": 0.2572012409567833,
|
|
"num_tokens": 20354697.0,
|
|
"step": 9830
|
|
},
|
|
{
|
|
"entropy": 5.3566303730010985,
|
|
"epoch": 8.933212176283508,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 6.49275209535242e-05,
|
|
"loss": 4.5272,
|
|
"mean_token_accuracy": 0.24859596490859986,
|
|
"num_tokens": 20365652.0,
|
|
"step": 9835
|
|
},
|
|
{
|
|
"entropy": 5.311676454544068,
|
|
"epoch": 8.937755565651976,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 6.480119159213288e-05,
|
|
"loss": 4.464,
|
|
"mean_token_accuracy": 0.2498265504837036,
|
|
"num_tokens": 20375629.0,
|
|
"step": 9840
|
|
},
|
|
{
|
|
"entropy": 5.309782886505127,
|
|
"epoch": 8.942298955020446,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 6.467538087540606e-05,
|
|
"loss": 4.513,
|
|
"mean_token_accuracy": 0.2444429039955139,
|
|
"num_tokens": 20385361.0,
|
|
"step": 9845
|
|
},
|
|
{
|
|
"entropy": 5.218897008895874,
|
|
"epoch": 8.946842344388914,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.455008911376914e-05,
|
|
"loss": 4.3791,
|
|
"mean_token_accuracy": 0.26221412122249604,
|
|
"num_tokens": 20394900.0,
|
|
"step": 9850
|
|
},
|
|
{
|
|
"entropy": 5.236014699935913,
|
|
"epoch": 8.951385733757384,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.442531661636715e-05,
|
|
"loss": 4.4538,
|
|
"mean_token_accuracy": 0.2497423157095909,
|
|
"num_tokens": 20405476.0,
|
|
"step": 9855
|
|
},
|
|
{
|
|
"entropy": 5.201627731323242,
|
|
"epoch": 8.955929123125852,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.430106369106381e-05,
|
|
"loss": 4.3866,
|
|
"mean_token_accuracy": 0.2632460817694664,
|
|
"num_tokens": 20414536.0,
|
|
"step": 9860
|
|
},
|
|
{
|
|
"entropy": 5.3509461879730225,
|
|
"epoch": 8.96047251249432,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 6.417733064444083e-05,
|
|
"loss": 4.5885,
|
|
"mean_token_accuracy": 0.24010050743818284,
|
|
"num_tokens": 20426046.0,
|
|
"step": 9865
|
|
},
|
|
{
|
|
"entropy": 5.205566024780273,
|
|
"epoch": 8.96501590186279,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.405411778179722e-05,
|
|
"loss": 4.3854,
|
|
"mean_token_accuracy": 0.26316569149494173,
|
|
"num_tokens": 20435788.0,
|
|
"step": 9870
|
|
},
|
|
{
|
|
"entropy": 5.247039794921875,
|
|
"epoch": 8.969559291231258,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.39314254071485e-05,
|
|
"loss": 4.4046,
|
|
"mean_token_accuracy": 0.26062908470630647,
|
|
"num_tokens": 20445166.0,
|
|
"step": 9875
|
|
},
|
|
{
|
|
"entropy": 5.220532178878784,
|
|
"epoch": 8.974102680599728,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.380925382322584e-05,
|
|
"loss": 4.4427,
|
|
"mean_token_accuracy": 0.25561596900224687,
|
|
"num_tokens": 20455034.0,
|
|
"step": 9880
|
|
},
|
|
{
|
|
"entropy": 5.270852708816529,
|
|
"epoch": 8.978646069968196,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.368760333147555e-05,
|
|
"loss": 4.4495,
|
|
"mean_token_accuracy": 0.2571917727589607,
|
|
"num_tokens": 20465092.0,
|
|
"step": 9885
|
|
},
|
|
{
|
|
"entropy": 5.277608442306518,
|
|
"epoch": 8.983189459336666,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.356647423205809e-05,
|
|
"loss": 4.4864,
|
|
"mean_token_accuracy": 0.24614014327526093,
|
|
"num_tokens": 20474793.0,
|
|
"step": 9890
|
|
},
|
|
{
|
|
"entropy": 5.234138107299804,
|
|
"epoch": 8.987732848705134,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 6.344586682384748e-05,
|
|
"loss": 4.4778,
|
|
"mean_token_accuracy": 0.2476370170712471,
|
|
"num_tokens": 20486201.0,
|
|
"step": 9895
|
|
},
|
|
{
|
|
"entropy": 5.262549161911011,
|
|
"epoch": 8.992276238073602,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 6.332578140443045e-05,
|
|
"loss": 4.5295,
|
|
"mean_token_accuracy": 0.24119538962841033,
|
|
"num_tokens": 20497264.0,
|
|
"step": 9900
|
|
},
|
|
{
|
|
"entropy": 5.240000772476196,
|
|
"epoch": 8.996819627442072,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 6.320621827010595e-05,
|
|
"loss": 4.4147,
|
|
"mean_token_accuracy": 0.26265947222709657,
|
|
"num_tokens": 20507279.0,
|
|
"step": 9905
|
|
},
|
|
{
|
|
"entropy": 5.255205207400852,
|
|
"epoch": 9.000908677873694,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.308717771588407e-05,
|
|
"loss": 4.3876,
|
|
"mean_token_accuracy": 0.25970009797149235,
|
|
"num_tokens": 20516472.0,
|
|
"step": 9910
|
|
},
|
|
{
|
|
"entropy": 5.248128461837768,
|
|
"epoch": 9.005452067242162,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.296866003548555e-05,
|
|
"loss": 4.4594,
|
|
"mean_token_accuracy": 0.2503984197974205,
|
|
"num_tokens": 20527251.0,
|
|
"step": 9915
|
|
},
|
|
{
|
|
"entropy": 5.255402088165283,
|
|
"epoch": 9.009995456610632,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.285066552134091e-05,
|
|
"loss": 4.3889,
|
|
"mean_token_accuracy": 0.25877925157547,
|
|
"num_tokens": 20537810.0,
|
|
"step": 9920
|
|
},
|
|
{
|
|
"entropy": 5.241429281234741,
|
|
"epoch": 9.0145388459791,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.273319446459002e-05,
|
|
"loss": 4.4091,
|
|
"mean_token_accuracy": 0.2563551217317581,
|
|
"num_tokens": 20548000.0,
|
|
"step": 9925
|
|
},
|
|
{
|
|
"entropy": 5.251785707473755,
|
|
"epoch": 9.01908223534757,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.261624715508095e-05,
|
|
"loss": 4.4136,
|
|
"mean_token_accuracy": 0.2606670454144478,
|
|
"num_tokens": 20557798.0,
|
|
"step": 9930
|
|
},
|
|
{
|
|
"entropy": 5.269507503509521,
|
|
"epoch": 9.023625624716038,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.249982388136958e-05,
|
|
"loss": 4.4572,
|
|
"mean_token_accuracy": 0.2559055030345917,
|
|
"num_tokens": 20568172.0,
|
|
"step": 9935
|
|
},
|
|
{
|
|
"entropy": 5.280324935913086,
|
|
"epoch": 9.028169014084508,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.238392493071873e-05,
|
|
"loss": 4.4264,
|
|
"mean_token_accuracy": 0.25376544445753096,
|
|
"num_tokens": 20578306.0,
|
|
"step": 9940
|
|
},
|
|
{
|
|
"entropy": 5.287593650817871,
|
|
"epoch": 9.032712403452976,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.226855058909762e-05,
|
|
"loss": 4.4203,
|
|
"mean_token_accuracy": 0.2620243728160858,
|
|
"num_tokens": 20588820.0,
|
|
"step": 9945
|
|
},
|
|
{
|
|
"entropy": 5.252154731750489,
|
|
"epoch": 9.037255792821444,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 6.215370114118092e-05,
|
|
"loss": 4.3813,
|
|
"mean_token_accuracy": 0.25924488008022306,
|
|
"num_tokens": 20598907.0,
|
|
"step": 9950
|
|
},
|
|
{
|
|
"entropy": 5.255476808547973,
|
|
"epoch": 9.041799182189914,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.203937687034822e-05,
|
|
"loss": 4.4646,
|
|
"mean_token_accuracy": 0.2523971349000931,
|
|
"num_tokens": 20609023.0,
|
|
"step": 9955
|
|
},
|
|
{
|
|
"entropy": 5.310302019119263,
|
|
"epoch": 9.046342571558382,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.192557805868328e-05,
|
|
"loss": 4.4375,
|
|
"mean_token_accuracy": 0.25857097953557967,
|
|
"num_tokens": 20619868.0,
|
|
"step": 9960
|
|
},
|
|
{
|
|
"entropy": 5.3225513935089115,
|
|
"epoch": 9.050885960926852,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.181230498697343e-05,
|
|
"loss": 4.4934,
|
|
"mean_token_accuracy": 0.2542723998427391,
|
|
"num_tokens": 20631674.0,
|
|
"step": 9965
|
|
},
|
|
{
|
|
"entropy": 5.310503005981445,
|
|
"epoch": 9.05542935029532,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.169955793470859e-05,
|
|
"loss": 4.4384,
|
|
"mean_token_accuracy": 0.2606181547045708,
|
|
"num_tokens": 20641878.0,
|
|
"step": 9970
|
|
},
|
|
{
|
|
"entropy": 5.307577657699585,
|
|
"epoch": 9.05997273966379,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.158733718008102e-05,
|
|
"loss": 4.4694,
|
|
"mean_token_accuracy": 0.25129154324531555,
|
|
"num_tokens": 20652663.0,
|
|
"step": 9975
|
|
},
|
|
{
|
|
"entropy": 5.242780876159668,
|
|
"epoch": 9.064516129032258,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.147564299998422e-05,
|
|
"loss": 4.4126,
|
|
"mean_token_accuracy": 0.2626940682530403,
|
|
"num_tokens": 20662651.0,
|
|
"step": 9980
|
|
},
|
|
{
|
|
"entropy": 5.248622751235962,
|
|
"epoch": 9.069059518400726,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.136447567001251e-05,
|
|
"loss": 4.4003,
|
|
"mean_token_accuracy": 0.2619339630007744,
|
|
"num_tokens": 20672531.0,
|
|
"step": 9985
|
|
},
|
|
{
|
|
"entropy": 5.255686473846436,
|
|
"epoch": 9.073602907769196,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 6.125383546446015e-05,
|
|
"loss": 4.4316,
|
|
"mean_token_accuracy": 0.2543242841959,
|
|
"num_tokens": 20681944.0,
|
|
"step": 9990
|
|
},
|
|
{
|
|
"entropy": 5.268953132629394,
|
|
"epoch": 9.078146297137664,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.114372265632095e-05,
|
|
"loss": 4.4179,
|
|
"mean_token_accuracy": 0.25196984112262727,
|
|
"num_tokens": 20692491.0,
|
|
"step": 9995
|
|
},
|
|
{
|
|
"entropy": 5.248694992065429,
|
|
"epoch": 9.082689686506134,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 6.1034137517287256e-05,
|
|
"loss": 4.3992,
|
|
"mean_token_accuracy": 0.26192291378974913,
|
|
"num_tokens": 20702649.0,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"epoch": 9.082689686506134,
|
|
"eval_entropy": 5.153491016875866,
|
|
"eval_loss": 5.676931858062744,
|
|
"eval_mean_token_accuracy": 0.1750378714917704,
|
|
"eval_num_tokens": 20702649.0,
|
|
"eval_runtime": 1.9994,
|
|
"eval_samples_per_second": 1718.026,
|
|
"eval_steps_per_second": 215.066,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"entropy": 5.268278551101685,
|
|
"epoch": 9.087233075874602,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.092508031774954e-05,
|
|
"loss": 4.3873,
|
|
"mean_token_accuracy": 0.2599007412791252,
|
|
"num_tokens": 20712543.0,
|
|
"step": 10005
|
|
},
|
|
{
|
|
"entropy": 5.232363224029541,
|
|
"epoch": 9.091776465243072,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.081655132679551e-05,
|
|
"loss": 4.3686,
|
|
"mean_token_accuracy": 0.2586787551641464,
|
|
"num_tokens": 20723500.0,
|
|
"step": 10010
|
|
},
|
|
{
|
|
"entropy": 5.239586257934571,
|
|
"epoch": 9.09631985461154,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 6.07085508122098e-05,
|
|
"loss": 4.3977,
|
|
"mean_token_accuracy": 0.2572086498141289,
|
|
"num_tokens": 20734141.0,
|
|
"step": 10015
|
|
},
|
|
{
|
|
"entropy": 5.260328865051269,
|
|
"epoch": 9.10086324398001,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.060107904047288e-05,
|
|
"loss": 4.3565,
|
|
"mean_token_accuracy": 0.2558779701590538,
|
|
"num_tokens": 20743486.0,
|
|
"step": 10020
|
|
},
|
|
{
|
|
"entropy": 5.2850462913513185,
|
|
"epoch": 9.105406633348478,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 6.0494136276760635e-05,
|
|
"loss": 4.4339,
|
|
"mean_token_accuracy": 0.2574521839618683,
|
|
"num_tokens": 20753577.0,
|
|
"step": 10025
|
|
},
|
|
{
|
|
"entropy": 5.291458320617676,
|
|
"epoch": 9.109950022716946,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.03877227849437e-05,
|
|
"loss": 4.4014,
|
|
"mean_token_accuracy": 0.2558879807591438,
|
|
"num_tokens": 20763829.0,
|
|
"step": 10030
|
|
},
|
|
{
|
|
"entropy": 5.252101039886474,
|
|
"epoch": 9.114493412085416,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.028183882758685e-05,
|
|
"loss": 4.3504,
|
|
"mean_token_accuracy": 0.26248141974210737,
|
|
"num_tokens": 20773450.0,
|
|
"step": 10035
|
|
},
|
|
{
|
|
"entropy": 5.2243914127349855,
|
|
"epoch": 9.119036801453884,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.01764846659482e-05,
|
|
"loss": 4.4112,
|
|
"mean_token_accuracy": 0.2567211419343948,
|
|
"num_tokens": 20783574.0,
|
|
"step": 10040
|
|
},
|
|
{
|
|
"entropy": 5.226731967926026,
|
|
"epoch": 9.123580190822354,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.0071660559978645e-05,
|
|
"loss": 4.3799,
|
|
"mean_token_accuracy": 0.2560179024934769,
|
|
"num_tokens": 20793130.0,
|
|
"step": 10045
|
|
},
|
|
{
|
|
"entropy": 5.271771430969238,
|
|
"epoch": 9.128123580190822,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.996736676832124e-05,
|
|
"loss": 4.4348,
|
|
"mean_token_accuracy": 0.25026746839284897,
|
|
"num_tokens": 20802766.0,
|
|
"step": 10050
|
|
},
|
|
{
|
|
"entropy": 5.241284894943237,
|
|
"epoch": 9.132666969559292,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.9863603548310646e-05,
|
|
"loss": 4.4263,
|
|
"mean_token_accuracy": 0.25605170875787736,
|
|
"num_tokens": 20812620.0,
|
|
"step": 10055
|
|
},
|
|
{
|
|
"entropy": 5.279885768890381,
|
|
"epoch": 9.13721035892776,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.976037115597215e-05,
|
|
"loss": 4.4181,
|
|
"mean_token_accuracy": 0.26014517396688464,
|
|
"num_tokens": 20822401.0,
|
|
"step": 10060
|
|
},
|
|
{
|
|
"entropy": 5.316309976577759,
|
|
"epoch": 9.141753748296228,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.965766984602149e-05,
|
|
"loss": 4.4864,
|
|
"mean_token_accuracy": 0.24782100021839143,
|
|
"num_tokens": 20832122.0,
|
|
"step": 10065
|
|
},
|
|
{
|
|
"entropy": 5.3360851287841795,
|
|
"epoch": 9.146297137664698,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 5.9555499871863955e-05,
|
|
"loss": 4.4654,
|
|
"mean_token_accuracy": 0.2509001776576042,
|
|
"num_tokens": 20843223.0,
|
|
"step": 10070
|
|
},
|
|
{
|
|
"entropy": 5.247539901733399,
|
|
"epoch": 9.150840527033166,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.9453861485593794e-05,
|
|
"loss": 4.4233,
|
|
"mean_token_accuracy": 0.2578518554568291,
|
|
"num_tokens": 20854452.0,
|
|
"step": 10075
|
|
},
|
|
{
|
|
"entropy": 5.292008113861084,
|
|
"epoch": 9.155383916401636,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.9352754937993586e-05,
|
|
"loss": 4.4519,
|
|
"mean_token_accuracy": 0.25286044776439665,
|
|
"num_tokens": 20864668.0,
|
|
"step": 10080
|
|
},
|
|
{
|
|
"entropy": 5.2490722179412845,
|
|
"epoch": 9.159927305770104,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.925218047853371e-05,
|
|
"loss": 4.4111,
|
|
"mean_token_accuracy": 0.2630553036928177,
|
|
"num_tokens": 20874667.0,
|
|
"step": 10085
|
|
},
|
|
{
|
|
"entropy": 5.241706275939942,
|
|
"epoch": 9.164470695138574,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.9152138355371646e-05,
|
|
"loss": 4.4776,
|
|
"mean_token_accuracy": 0.257026407122612,
|
|
"num_tokens": 20884578.0,
|
|
"step": 10090
|
|
},
|
|
{
|
|
"entropy": 5.279536867141724,
|
|
"epoch": 9.169014084507042,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.9052628815351416e-05,
|
|
"loss": 4.4284,
|
|
"mean_token_accuracy": 0.25599348694086077,
|
|
"num_tokens": 20894239.0,
|
|
"step": 10095
|
|
},
|
|
{
|
|
"entropy": 5.3282524108886715,
|
|
"epoch": 9.173557473875512,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.8953652104002825e-05,
|
|
"loss": 4.3897,
|
|
"mean_token_accuracy": 0.2617577463388443,
|
|
"num_tokens": 20904960.0,
|
|
"step": 10100
|
|
},
|
|
{
|
|
"entropy": 5.278601264953613,
|
|
"epoch": 9.17810086324398,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.885520846554119e-05,
|
|
"loss": 4.4336,
|
|
"mean_token_accuracy": 0.2535366639494896,
|
|
"num_tokens": 20915215.0,
|
|
"step": 10105
|
|
},
|
|
{
|
|
"entropy": 5.266632223129273,
|
|
"epoch": 9.182644252612448,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.875729814286634e-05,
|
|
"loss": 4.4235,
|
|
"mean_token_accuracy": 0.2597662076354027,
|
|
"num_tokens": 20925112.0,
|
|
"step": 10110
|
|
},
|
|
{
|
|
"entropy": 5.2537269115448,
|
|
"epoch": 9.187187641980918,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.865992137756225e-05,
|
|
"loss": 4.4064,
|
|
"mean_token_accuracy": 0.25972014367580415,
|
|
"num_tokens": 20935393.0,
|
|
"step": 10115
|
|
},
|
|
{
|
|
"entropy": 5.274113702774048,
|
|
"epoch": 9.191731031349386,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.856307840989642e-05,
|
|
"loss": 4.461,
|
|
"mean_token_accuracy": 0.2506603911519051,
|
|
"num_tokens": 20944876.0,
|
|
"step": 10120
|
|
},
|
|
{
|
|
"entropy": 5.258919811248779,
|
|
"epoch": 9.196274420717856,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.846676947881927e-05,
|
|
"loss": 4.4253,
|
|
"mean_token_accuracy": 0.2557604253292084,
|
|
"num_tokens": 20955115.0,
|
|
"step": 10125
|
|
},
|
|
{
|
|
"entropy": 5.25908932685852,
|
|
"epoch": 9.200817810086324,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.8370994821963514e-05,
|
|
"loss": 4.4601,
|
|
"mean_token_accuracy": 0.2509754791855812,
|
|
"num_tokens": 20964914.0,
|
|
"step": 10130
|
|
},
|
|
{
|
|
"entropy": 5.275650548934936,
|
|
"epoch": 9.205361199454794,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.8275754675643565e-05,
|
|
"loss": 4.3818,
|
|
"mean_token_accuracy": 0.2593634456396103,
|
|
"num_tokens": 20975768.0,
|
|
"step": 10135
|
|
},
|
|
{
|
|
"entropy": 5.224311971664429,
|
|
"epoch": 9.209904588823262,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.8181049274855013e-05,
|
|
"loss": 4.3814,
|
|
"mean_token_accuracy": 0.2575690194964409,
|
|
"num_tokens": 20985496.0,
|
|
"step": 10140
|
|
},
|
|
{
|
|
"entropy": 5.308774042129516,
|
|
"epoch": 9.214447978191732,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.80868788532741e-05,
|
|
"loss": 4.4736,
|
|
"mean_token_accuracy": 0.25429400354623793,
|
|
"num_tokens": 20995243.0,
|
|
"step": 10145
|
|
},
|
|
{
|
|
"entropy": 5.228246021270752,
|
|
"epoch": 9.2189913675602,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.7993243643256854e-05,
|
|
"loss": 4.4606,
|
|
"mean_token_accuracy": 0.2507553219795227,
|
|
"num_tokens": 21005768.0,
|
|
"step": 10150
|
|
},
|
|
{
|
|
"entropy": 5.304186868667602,
|
|
"epoch": 9.223534756928668,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.790014387583895e-05,
|
|
"loss": 4.5066,
|
|
"mean_token_accuracy": 0.24594873934984207,
|
|
"num_tokens": 21015615.0,
|
|
"step": 10155
|
|
},
|
|
{
|
|
"entropy": 5.309623908996582,
|
|
"epoch": 9.228078146297138,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.780757978073474e-05,
|
|
"loss": 4.4879,
|
|
"mean_token_accuracy": 0.2529712840914726,
|
|
"num_tokens": 21026482.0,
|
|
"step": 10160
|
|
},
|
|
{
|
|
"entropy": 5.292509460449219,
|
|
"epoch": 9.232621535665606,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.771555158633699e-05,
|
|
"loss": 4.4984,
|
|
"mean_token_accuracy": 0.24785834550857544,
|
|
"num_tokens": 21037709.0,
|
|
"step": 10165
|
|
},
|
|
{
|
|
"entropy": 5.248999834060669,
|
|
"epoch": 9.237164925034076,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.762405951971607e-05,
|
|
"loss": 4.341,
|
|
"mean_token_accuracy": 0.26388682425022125,
|
|
"num_tokens": 21047306.0,
|
|
"step": 10170
|
|
},
|
|
{
|
|
"entropy": 5.241924381256103,
|
|
"epoch": 9.241708314402544,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.75331038066196e-05,
|
|
"loss": 4.3965,
|
|
"mean_token_accuracy": 0.2633807063102722,
|
|
"num_tokens": 21057271.0,
|
|
"step": 10175
|
|
},
|
|
{
|
|
"entropy": 5.310989665985107,
|
|
"epoch": 9.246251703771014,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 5.744268467147174e-05,
|
|
"loss": 4.4439,
|
|
"mean_token_accuracy": 0.24884820133447647,
|
|
"num_tokens": 21066404.0,
|
|
"step": 10180
|
|
},
|
|
{
|
|
"entropy": 5.278117513656616,
|
|
"epoch": 9.250795093139482,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.7352802337372724e-05,
|
|
"loss": 4.4302,
|
|
"mean_token_accuracy": 0.2511978641152382,
|
|
"num_tokens": 21077101.0,
|
|
"step": 10185
|
|
},
|
|
{
|
|
"entropy": 5.385194301605225,
|
|
"epoch": 9.25533848250795,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.726345702609824e-05,
|
|
"loss": 4.5542,
|
|
"mean_token_accuracy": 0.24398215264081954,
|
|
"num_tokens": 21087640.0,
|
|
"step": 10190
|
|
},
|
|
{
|
|
"entropy": 5.1991735935211185,
|
|
"epoch": 9.25988187187642,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.7174648958099e-05,
|
|
"loss": 4.3268,
|
|
"mean_token_accuracy": 0.26411201506853105,
|
|
"num_tokens": 21098394.0,
|
|
"step": 10195
|
|
},
|
|
{
|
|
"entropy": 5.287927436828613,
|
|
"epoch": 9.264425261244888,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 5.7086378352500096e-05,
|
|
"loss": 4.4737,
|
|
"mean_token_accuracy": 0.2515187978744507,
|
|
"num_tokens": 21108949.0,
|
|
"step": 10200
|
|
},
|
|
{
|
|
"entropy": 5.28193941116333,
|
|
"epoch": 9.268968650613358,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.6998645427100456e-05,
|
|
"loss": 4.464,
|
|
"mean_token_accuracy": 0.25657002776861193,
|
|
"num_tokens": 21119204.0,
|
|
"step": 10205
|
|
},
|
|
{
|
|
"entropy": 5.24984884262085,
|
|
"epoch": 9.273512039981826,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.691145039837231e-05,
|
|
"loss": 4.357,
|
|
"mean_token_accuracy": 0.26669247448444366,
|
|
"num_tokens": 21128959.0,
|
|
"step": 10210
|
|
},
|
|
{
|
|
"entropy": 5.271595096588134,
|
|
"epoch": 9.278055429350296,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 5.682479348146081e-05,
|
|
"loss": 4.4053,
|
|
"mean_token_accuracy": 0.2523984402418137,
|
|
"num_tokens": 21139465.0,
|
|
"step": 10215
|
|
},
|
|
{
|
|
"entropy": 5.296364259719849,
|
|
"epoch": 9.282598818718764,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.673867489018322e-05,
|
|
"loss": 4.4521,
|
|
"mean_token_accuracy": 0.2582927107810974,
|
|
"num_tokens": 21150871.0,
|
|
"step": 10220
|
|
},
|
|
{
|
|
"entropy": 5.249798536300659,
|
|
"epoch": 9.287142208087234,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.6653094837028626e-05,
|
|
"loss": 4.418,
|
|
"mean_token_accuracy": 0.2573807165026665,
|
|
"num_tokens": 21161905.0,
|
|
"step": 10225
|
|
},
|
|
{
|
|
"entropy": 5.202979660034179,
|
|
"epoch": 9.291685597455702,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.6568053533157275e-05,
|
|
"loss": 4.3734,
|
|
"mean_token_accuracy": 0.2592195361852646,
|
|
"num_tokens": 21172367.0,
|
|
"step": 10230
|
|
},
|
|
{
|
|
"entropy": 5.330692005157471,
|
|
"epoch": 9.29622898682417,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 5.648355118840018e-05,
|
|
"loss": 4.5102,
|
|
"mean_token_accuracy": 0.24727253913879393,
|
|
"num_tokens": 21182471.0,
|
|
"step": 10235
|
|
},
|
|
{
|
|
"entropy": 5.275740671157837,
|
|
"epoch": 9.30077237619264,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.639958801125842e-05,
|
|
"loss": 4.4653,
|
|
"mean_token_accuracy": 0.25160403549671173,
|
|
"num_tokens": 21193292.0,
|
|
"step": 10240
|
|
},
|
|
{
|
|
"entropy": 5.266654109954834,
|
|
"epoch": 9.305315765561108,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.6316164208902844e-05,
|
|
"loss": 4.3745,
|
|
"mean_token_accuracy": 0.2618174538016319,
|
|
"num_tokens": 21203711.0,
|
|
"step": 10245
|
|
},
|
|
{
|
|
"entropy": 5.293562650680542,
|
|
"epoch": 9.309859154929578,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.623327998717334e-05,
|
|
"loss": 4.4484,
|
|
"mean_token_accuracy": 0.2563167795538902,
|
|
"num_tokens": 21214037.0,
|
|
"step": 10250
|
|
},
|
|
{
|
|
"entropy": 5.320832777023315,
|
|
"epoch": 9.314402544298046,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.615093555057853e-05,
|
|
"loss": 4.5018,
|
|
"mean_token_accuracy": 0.24982865005731583,
|
|
"num_tokens": 21225514.0,
|
|
"step": 10255
|
|
},
|
|
{
|
|
"entropy": 5.254089212417602,
|
|
"epoch": 9.318945933666516,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.606913110229511e-05,
|
|
"loss": 4.4299,
|
|
"mean_token_accuracy": 0.25660114735364914,
|
|
"num_tokens": 21236002.0,
|
|
"step": 10260
|
|
},
|
|
{
|
|
"entropy": 5.3174420356750485,
|
|
"epoch": 9.323489323034984,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.598786684416738e-05,
|
|
"loss": 4.493,
|
|
"mean_token_accuracy": 0.2571220889687538,
|
|
"num_tokens": 21245785.0,
|
|
"step": 10265
|
|
},
|
|
{
|
|
"entropy": 5.290515041351318,
|
|
"epoch": 9.328032712403452,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.59071429767069e-05,
|
|
"loss": 4.451,
|
|
"mean_token_accuracy": 0.2535445377230644,
|
|
"num_tokens": 21255396.0,
|
|
"step": 10270
|
|
},
|
|
{
|
|
"entropy": 5.318304347991943,
|
|
"epoch": 9.332576101771922,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.5826959699091725e-05,
|
|
"loss": 4.4555,
|
|
"mean_token_accuracy": 0.2603874370455742,
|
|
"num_tokens": 21265341.0,
|
|
"step": 10275
|
|
},
|
|
{
|
|
"entropy": 5.242697715759277,
|
|
"epoch": 9.33711949114039,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.574731720916617e-05,
|
|
"loss": 4.4389,
|
|
"mean_token_accuracy": 0.24402630925178528,
|
|
"num_tokens": 21276344.0,
|
|
"step": 10280
|
|
},
|
|
{
|
|
"entropy": 5.331905698776245,
|
|
"epoch": 9.34166288050886,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.56682157034401e-05,
|
|
"loss": 4.4362,
|
|
"mean_token_accuracy": 0.2528820544481277,
|
|
"num_tokens": 21286836.0,
|
|
"step": 10285
|
|
},
|
|
{
|
|
"entropy": 5.282867050170898,
|
|
"epoch": 9.346206269877328,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.558965537708867e-05,
|
|
"loss": 4.4008,
|
|
"mean_token_accuracy": 0.2558562561869621,
|
|
"num_tokens": 21297064.0,
|
|
"step": 10290
|
|
},
|
|
{
|
|
"entropy": 5.286652326583862,
|
|
"epoch": 9.350749659245798,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 5.551163642395165e-05,
|
|
"loss": 4.4357,
|
|
"mean_token_accuracy": 0.26118839234113694,
|
|
"num_tokens": 21307687.0,
|
|
"step": 10295
|
|
},
|
|
{
|
|
"entropy": 5.206988382339477,
|
|
"epoch": 9.355293048614266,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.543415903653307e-05,
|
|
"loss": 4.4231,
|
|
"mean_token_accuracy": 0.2540373563766479,
|
|
"num_tokens": 21317596.0,
|
|
"step": 10300
|
|
},
|
|
{
|
|
"entropy": 5.276752710342407,
|
|
"epoch": 9.359836437982736,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 5.535722340600065e-05,
|
|
"loss": 4.4742,
|
|
"mean_token_accuracy": 0.25186550468206403,
|
|
"num_tokens": 21327515.0,
|
|
"step": 10305
|
|
},
|
|
{
|
|
"entropy": 5.307286643981934,
|
|
"epoch": 9.364379827351204,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.528082972218548e-05,
|
|
"loss": 4.5336,
|
|
"mean_token_accuracy": 0.24522192031145096,
|
|
"num_tokens": 21338583.0,
|
|
"step": 10310
|
|
},
|
|
{
|
|
"entropy": 5.303515768051147,
|
|
"epoch": 9.368923216719672,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 5.520497817358129e-05,
|
|
"loss": 4.5082,
|
|
"mean_token_accuracy": 0.2503837287425995,
|
|
"num_tokens": 21349119.0,
|
|
"step": 10315
|
|
},
|
|
{
|
|
"entropy": 5.347152519226074,
|
|
"epoch": 9.373466606088142,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 5.5129668947344275e-05,
|
|
"loss": 4.4875,
|
|
"mean_token_accuracy": 0.2591558337211609,
|
|
"num_tokens": 21359997.0,
|
|
"step": 10320
|
|
},
|
|
{
|
|
"entropy": 5.277323198318482,
|
|
"epoch": 9.37800999545661,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 5.505490222929246e-05,
|
|
"loss": 4.4323,
|
|
"mean_token_accuracy": 0.25226314961910246,
|
|
"num_tokens": 21369457.0,
|
|
"step": 10325
|
|
},
|
|
{
|
|
"entropy": 5.3224297046661375,
|
|
"epoch": 9.38255338482508,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.498067820390537e-05,
|
|
"loss": 4.5053,
|
|
"mean_token_accuracy": 0.2461426466703415,
|
|
"num_tokens": 21380230.0,
|
|
"step": 10330
|
|
},
|
|
{
|
|
"entropy": 5.219454336166382,
|
|
"epoch": 9.387096774193548,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.4906997054323245e-05,
|
|
"loss": 4.4063,
|
|
"mean_token_accuracy": 0.26092179864645004,
|
|
"num_tokens": 21390235.0,
|
|
"step": 10335
|
|
},
|
|
{
|
|
"entropy": 5.25851035118103,
|
|
"epoch": 9.391640163562018,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.483385896234714e-05,
|
|
"loss": 4.432,
|
|
"mean_token_accuracy": 0.2595922589302063,
|
|
"num_tokens": 21400034.0,
|
|
"step": 10340
|
|
},
|
|
{
|
|
"entropy": 5.259699487686158,
|
|
"epoch": 9.396183552930486,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.4761264108437966e-05,
|
|
"loss": 4.392,
|
|
"mean_token_accuracy": 0.25419065207242963,
|
|
"num_tokens": 21410136.0,
|
|
"step": 10345
|
|
},
|
|
{
|
|
"entropy": 5.2928050518035885,
|
|
"epoch": 9.400726942298956,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.4689212671716335e-05,
|
|
"loss": 4.4643,
|
|
"mean_token_accuracy": 0.25648984462022784,
|
|
"num_tokens": 21420333.0,
|
|
"step": 10350
|
|
},
|
|
{
|
|
"entropy": 5.275780820846558,
|
|
"epoch": 9.405270331667424,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.461770482996194e-05,
|
|
"loss": 4.4328,
|
|
"mean_token_accuracy": 0.25249586552381514,
|
|
"num_tokens": 21430804.0,
|
|
"step": 10355
|
|
},
|
|
{
|
|
"entropy": 5.2289087772369385,
|
|
"epoch": 9.409813721035892,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.4546740759613354e-05,
|
|
"loss": 4.4044,
|
|
"mean_token_accuracy": 0.25471530854701996,
|
|
"num_tokens": 21440769.0,
|
|
"step": 10360
|
|
},
|
|
{
|
|
"entropy": 5.324357652664185,
|
|
"epoch": 9.414357110404362,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.447632063576734e-05,
|
|
"loss": 4.4604,
|
|
"mean_token_accuracy": 0.25341477394104006,
|
|
"num_tokens": 21450648.0,
|
|
"step": 10365
|
|
},
|
|
{
|
|
"entropy": 5.28548526763916,
|
|
"epoch": 9.41890049977283,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.4406444632178515e-05,
|
|
"loss": 4.4618,
|
|
"mean_token_accuracy": 0.25136143416166307,
|
|
"num_tokens": 21461038.0,
|
|
"step": 10370
|
|
},
|
|
{
|
|
"entropy": 5.297026443481445,
|
|
"epoch": 9.4234438891413,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.4337112921259005e-05,
|
|
"loss": 4.4535,
|
|
"mean_token_accuracy": 0.25314539968967437,
|
|
"num_tokens": 21471702.0,
|
|
"step": 10375
|
|
},
|
|
{
|
|
"entropy": 5.294500255584717,
|
|
"epoch": 9.427987278509768,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.426832567407794e-05,
|
|
"loss": 4.3914,
|
|
"mean_token_accuracy": 0.2555156469345093,
|
|
"num_tokens": 21482434.0,
|
|
"step": 10380
|
|
},
|
|
{
|
|
"entropy": 5.265732622146606,
|
|
"epoch": 9.432530667878238,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.420008306036098e-05,
|
|
"loss": 4.4078,
|
|
"mean_token_accuracy": 0.2603926002979279,
|
|
"num_tokens": 21492738.0,
|
|
"step": 10385
|
|
},
|
|
{
|
|
"entropy": 5.245393991470337,
|
|
"epoch": 9.437074057246706,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.413238524848998e-05,
|
|
"loss": 4.4124,
|
|
"mean_token_accuracy": 0.2581202477216721,
|
|
"num_tokens": 21503154.0,
|
|
"step": 10390
|
|
},
|
|
{
|
|
"entropy": 5.274657249450684,
|
|
"epoch": 9.441617446615174,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.4065232405502576e-05,
|
|
"loss": 4.4596,
|
|
"mean_token_accuracy": 0.24760989099740982,
|
|
"num_tokens": 21513097.0,
|
|
"step": 10395
|
|
},
|
|
{
|
|
"entropy": 5.31263370513916,
|
|
"epoch": 9.446160835983644,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 5.399862469709175e-05,
|
|
"loss": 4.4501,
|
|
"mean_token_accuracy": 0.25185920894145963,
|
|
"num_tokens": 21522888.0,
|
|
"step": 10400
|
|
},
|
|
{
|
|
"entropy": 5.2880888938903805,
|
|
"epoch": 9.450704225352112,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 5.3932562287605364e-05,
|
|
"loss": 4.4048,
|
|
"mean_token_accuracy": 0.2588182032108307,
|
|
"num_tokens": 21533998.0,
|
|
"step": 10405
|
|
},
|
|
{
|
|
"entropy": 5.229194974899292,
|
|
"epoch": 9.455247614720582,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.38670453400459e-05,
|
|
"loss": 4.4913,
|
|
"mean_token_accuracy": 0.2485425814986229,
|
|
"num_tokens": 21544042.0,
|
|
"step": 10410
|
|
},
|
|
{
|
|
"entropy": 5.209096527099609,
|
|
"epoch": 9.45979100408905,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 5.3802074016069835e-05,
|
|
"loss": 4.3898,
|
|
"mean_token_accuracy": 0.26366556733846663,
|
|
"num_tokens": 21553449.0,
|
|
"step": 10415
|
|
},
|
|
{
|
|
"entropy": 5.296755599975586,
|
|
"epoch": 9.46433439345752,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 5.373764847598756e-05,
|
|
"loss": 4.4795,
|
|
"mean_token_accuracy": 0.24844876527786255,
|
|
"num_tokens": 21564485.0,
|
|
"step": 10420
|
|
},
|
|
{
|
|
"entropy": 5.334478807449341,
|
|
"epoch": 9.468877782825988,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.3673768878762564e-05,
|
|
"loss": 4.552,
|
|
"mean_token_accuracy": 0.24421423375606538,
|
|
"num_tokens": 21574889.0,
|
|
"step": 10425
|
|
},
|
|
{
|
|
"entropy": 5.233843755722046,
|
|
"epoch": 9.473421172194458,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 5.361043538201152e-05,
|
|
"loss": 4.4087,
|
|
"mean_token_accuracy": 0.2653982609510422,
|
|
"num_tokens": 21584913.0,
|
|
"step": 10430
|
|
},
|
|
{
|
|
"entropy": 5.249214696884155,
|
|
"epoch": 9.477964561562926,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 5.3547648142003454e-05,
|
|
"loss": 4.3614,
|
|
"mean_token_accuracy": 0.2629379898309708,
|
|
"num_tokens": 21594711.0,
|
|
"step": 10435
|
|
},
|
|
{
|
|
"entropy": 5.241414785385132,
|
|
"epoch": 9.482507950931394,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.348540731365968e-05,
|
|
"loss": 4.3751,
|
|
"mean_token_accuracy": 0.26117991358041764,
|
|
"num_tokens": 21605100.0,
|
|
"step": 10440
|
|
},
|
|
{
|
|
"entropy": 5.264820623397827,
|
|
"epoch": 9.487051340299864,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.342371305055323e-05,
|
|
"loss": 4.4562,
|
|
"mean_token_accuracy": 0.2573784992098808,
|
|
"num_tokens": 21616153.0,
|
|
"step": 10445
|
|
},
|
|
{
|
|
"entropy": 5.301622581481934,
|
|
"epoch": 9.491594729668332,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 5.3362565504908595e-05,
|
|
"loss": 4.4714,
|
|
"mean_token_accuracy": 0.2634389892220497,
|
|
"num_tokens": 21626102.0,
|
|
"step": 10450
|
|
},
|
|
{
|
|
"entropy": 5.213448238372803,
|
|
"epoch": 9.496138119036802,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.330196482760126e-05,
|
|
"loss": 4.3471,
|
|
"mean_token_accuracy": 0.26391187906265257,
|
|
"num_tokens": 21635338.0,
|
|
"step": 10455
|
|
},
|
|
{
|
|
"entropy": 5.3183294296264645,
|
|
"epoch": 9.50068150840527,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.324191116815735e-05,
|
|
"loss": 4.4911,
|
|
"mean_token_accuracy": 0.2543415635824203,
|
|
"num_tokens": 21645783.0,
|
|
"step": 10460
|
|
},
|
|
{
|
|
"entropy": 5.215113639831543,
|
|
"epoch": 9.50522489777374,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.31824046747533e-05,
|
|
"loss": 4.3711,
|
|
"mean_token_accuracy": 0.26015513986349104,
|
|
"num_tokens": 21656112.0,
|
|
"step": 10465
|
|
},
|
|
{
|
|
"entropy": 5.317779874801635,
|
|
"epoch": 9.509768287142208,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.312344549421547e-05,
|
|
"loss": 4.4253,
|
|
"mean_token_accuracy": 0.258123454451561,
|
|
"num_tokens": 21667886.0,
|
|
"step": 10470
|
|
},
|
|
{
|
|
"entropy": 5.269112539291382,
|
|
"epoch": 9.514311676510676,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.306503377201981e-05,
|
|
"loss": 4.3837,
|
|
"mean_token_accuracy": 0.26316780894994735,
|
|
"num_tokens": 21677726.0,
|
|
"step": 10475
|
|
},
|
|
{
|
|
"entropy": 5.265356254577637,
|
|
"epoch": 9.518855065879146,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.300716965229141e-05,
|
|
"loss": 4.4799,
|
|
"mean_token_accuracy": 0.25488703697919846,
|
|
"num_tokens": 21688552.0,
|
|
"step": 10480
|
|
},
|
|
{
|
|
"entropy": 5.277896451950073,
|
|
"epoch": 9.523398455247614,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.294985327780423e-05,
|
|
"loss": 4.4358,
|
|
"mean_token_accuracy": 0.25569491684436796,
|
|
"num_tokens": 21699206.0,
|
|
"step": 10485
|
|
},
|
|
{
|
|
"entropy": 5.279299688339234,
|
|
"epoch": 9.527941844616084,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.289308478998076e-05,
|
|
"loss": 4.3685,
|
|
"mean_token_accuracy": 0.26114382445812223,
|
|
"num_tokens": 21709083.0,
|
|
"step": 10490
|
|
},
|
|
{
|
|
"entropy": 5.209085988998413,
|
|
"epoch": 9.532485233984552,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.2836864328891533e-05,
|
|
"loss": 4.407,
|
|
"mean_token_accuracy": 0.254348586499691,
|
|
"num_tokens": 21719092.0,
|
|
"step": 10495
|
|
},
|
|
{
|
|
"entropy": 5.226184940338134,
|
|
"epoch": 9.537028623353022,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.278119203325501e-05,
|
|
"loss": 4.4803,
|
|
"mean_token_accuracy": 0.2512966737151146,
|
|
"num_tokens": 21729335.0,
|
|
"step": 10500
|
|
},
|
|
{
|
|
"epoch": 9.537028623353022,
|
|
"eval_entropy": 5.144930997005729,
|
|
"eval_loss": 5.67848014831543,
|
|
"eval_mean_token_accuracy": 0.17503768749708354,
|
|
"eval_num_tokens": 21729335.0,
|
|
"eval_runtime": 2.1913,
|
|
"eval_samples_per_second": 1567.554,
|
|
"eval_steps_per_second": 196.229,
|
|
"step": 10500
|
|
},
|
|
{
|
|
"entropy": 5.241572856903076,
|
|
"epoch": 9.54157201272149,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 5.272606804043701e-05,
|
|
"loss": 4.4339,
|
|
"mean_token_accuracy": 0.2583494037389755,
|
|
"num_tokens": 21740135.0,
|
|
"step": 10505
|
|
},
|
|
{
|
|
"entropy": 5.223893451690674,
|
|
"epoch": 9.54611540208996,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.267149248645054e-05,
|
|
"loss": 4.3169,
|
|
"mean_token_accuracy": 0.2662604734301567,
|
|
"num_tokens": 21749481.0,
|
|
"step": 10510
|
|
},
|
|
{
|
|
"entropy": 5.311245918273926,
|
|
"epoch": 9.550658791458428,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 5.2617465505955324e-05,
|
|
"loss": 4.5284,
|
|
"mean_token_accuracy": 0.24050216972827912,
|
|
"num_tokens": 21760175.0,
|
|
"step": 10515
|
|
},
|
|
{
|
|
"entropy": 5.258873844146729,
|
|
"epoch": 9.555202180826896,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.256398723225758e-05,
|
|
"loss": 4.415,
|
|
"mean_token_accuracy": 0.26078986525535586,
|
|
"num_tokens": 21771138.0,
|
|
"step": 10520
|
|
},
|
|
{
|
|
"entropy": 5.222827672958374,
|
|
"epoch": 9.559745570195366,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 5.251105779730963e-05,
|
|
"loss": 4.4523,
|
|
"mean_token_accuracy": 0.2554008632898331,
|
|
"num_tokens": 21781459.0,
|
|
"step": 10525
|
|
},
|
|
{
|
|
"entropy": 5.209052038192749,
|
|
"epoch": 9.564288959563834,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.2458677331709585e-05,
|
|
"loss": 4.413,
|
|
"mean_token_accuracy": 0.26385961621999743,
|
|
"num_tokens": 21792265.0,
|
|
"step": 10530
|
|
},
|
|
{
|
|
"entropy": 5.280862617492676,
|
|
"epoch": 9.568832348932304,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.240684596470104e-05,
|
|
"loss": 4.4775,
|
|
"mean_token_accuracy": 0.24692262560129166,
|
|
"num_tokens": 21802379.0,
|
|
"step": 10535
|
|
},
|
|
{
|
|
"entropy": 5.27555742263794,
|
|
"epoch": 9.573375738300772,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 5.2355563824172744e-05,
|
|
"loss": 4.4948,
|
|
"mean_token_accuracy": 0.24436188787221907,
|
|
"num_tokens": 21812910.0,
|
|
"step": 10540
|
|
},
|
|
{
|
|
"entropy": 5.271632242202759,
|
|
"epoch": 9.577919127669242,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.23048310366583e-05,
|
|
"loss": 4.4715,
|
|
"mean_token_accuracy": 0.25190804451704024,
|
|
"num_tokens": 21823619.0,
|
|
"step": 10545
|
|
},
|
|
{
|
|
"entropy": 5.31701693534851,
|
|
"epoch": 9.58246251703771,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.2254647727335795e-05,
|
|
"loss": 4.5573,
|
|
"mean_token_accuracy": 0.24523578584194183,
|
|
"num_tokens": 21833661.0,
|
|
"step": 10550
|
|
},
|
|
{
|
|
"entropy": 5.23261342048645,
|
|
"epoch": 9.58700590640618,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.220501402002755e-05,
|
|
"loss": 4.3646,
|
|
"mean_token_accuracy": 0.2565658986568451,
|
|
"num_tokens": 21843528.0,
|
|
"step": 10555
|
|
},
|
|
{
|
|
"entropy": 5.246652460098266,
|
|
"epoch": 9.591549295774648,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.215593003719981e-05,
|
|
"loss": 4.3978,
|
|
"mean_token_accuracy": 0.25955182164907453,
|
|
"num_tokens": 21854282.0,
|
|
"step": 10560
|
|
},
|
|
{
|
|
"entropy": 5.275947141647339,
|
|
"epoch": 9.596092685143116,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.210739589996245e-05,
|
|
"loss": 4.458,
|
|
"mean_token_accuracy": 0.2508491098880768,
|
|
"num_tokens": 21864631.0,
|
|
"step": 10565
|
|
},
|
|
{
|
|
"entropy": 5.227098846435547,
|
|
"epoch": 9.600636074511586,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.2059411728068595e-05,
|
|
"loss": 4.4273,
|
|
"mean_token_accuracy": 0.25220730155706406,
|
|
"num_tokens": 21875298.0,
|
|
"step": 10570
|
|
},
|
|
{
|
|
"entropy": 5.297574234008789,
|
|
"epoch": 9.605179463880054,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.201197763991444e-05,
|
|
"loss": 4.4891,
|
|
"mean_token_accuracy": 0.2457113519310951,
|
|
"num_tokens": 21886317.0,
|
|
"step": 10575
|
|
},
|
|
{
|
|
"entropy": 5.302400875091553,
|
|
"epoch": 9.609722853248524,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.196509375253887e-05,
|
|
"loss": 4.4366,
|
|
"mean_token_accuracy": 0.2555605500936508,
|
|
"num_tokens": 21897459.0,
|
|
"step": 10580
|
|
},
|
|
{
|
|
"entropy": 5.307972621917725,
|
|
"epoch": 9.614266242616992,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.19187601816232e-05,
|
|
"loss": 4.4514,
|
|
"mean_token_accuracy": 0.2527316227555275,
|
|
"num_tokens": 21909122.0,
|
|
"step": 10585
|
|
},
|
|
{
|
|
"entropy": 5.269607353210449,
|
|
"epoch": 9.618809631985462,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.187297704149094e-05,
|
|
"loss": 4.3918,
|
|
"mean_token_accuracy": 0.26127581149339674,
|
|
"num_tokens": 21918610.0,
|
|
"step": 10590
|
|
},
|
|
{
|
|
"entropy": 5.289934778213501,
|
|
"epoch": 9.62335302135393,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 5.182774444510744e-05,
|
|
"loss": 4.3971,
|
|
"mean_token_accuracy": 0.2662158638238907,
|
|
"num_tokens": 21927697.0,
|
|
"step": 10595
|
|
},
|
|
{
|
|
"entropy": 5.253153276443482,
|
|
"epoch": 9.627896410722398,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.178306250407963e-05,
|
|
"loss": 4.3698,
|
|
"mean_token_accuracy": 0.26760995388031006,
|
|
"num_tokens": 21936750.0,
|
|
"step": 10600
|
|
},
|
|
{
|
|
"entropy": 5.261616992950439,
|
|
"epoch": 9.632439800090868,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.1738931328655714e-05,
|
|
"loss": 4.3998,
|
|
"mean_token_accuracy": 0.26440925896167755,
|
|
"num_tokens": 21946291.0,
|
|
"step": 10605
|
|
},
|
|
{
|
|
"entropy": 5.251005792617798,
|
|
"epoch": 9.636983189459336,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 5.1695351027725054e-05,
|
|
"loss": 4.3877,
|
|
"mean_token_accuracy": 0.264467166364193,
|
|
"num_tokens": 21956102.0,
|
|
"step": 10610
|
|
},
|
|
{
|
|
"entropy": 5.236642932891845,
|
|
"epoch": 9.641526578827806,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.1652321708817646e-05,
|
|
"loss": 4.4038,
|
|
"mean_token_accuracy": 0.2568131759762764,
|
|
"num_tokens": 21966953.0,
|
|
"step": 10615
|
|
},
|
|
{
|
|
"entropy": 5.327154350280762,
|
|
"epoch": 9.646069968196274,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.160984347810409e-05,
|
|
"loss": 4.4894,
|
|
"mean_token_accuracy": 0.2554873988032341,
|
|
"num_tokens": 21977461.0,
|
|
"step": 10620
|
|
},
|
|
{
|
|
"entropy": 5.304117107391358,
|
|
"epoch": 9.650613357564744,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.15679164403952e-05,
|
|
"loss": 4.4983,
|
|
"mean_token_accuracy": 0.24222443848848343,
|
|
"num_tokens": 21988571.0,
|
|
"step": 10625
|
|
},
|
|
{
|
|
"entropy": 5.328456163406372,
|
|
"epoch": 9.655156746933212,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.152654069914177e-05,
|
|
"loss": 4.4732,
|
|
"mean_token_accuracy": 0.255735944211483,
|
|
"num_tokens": 22000503.0,
|
|
"step": 10630
|
|
},
|
|
{
|
|
"entropy": 5.227297973632813,
|
|
"epoch": 9.65970013630168,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.1485716356434315e-05,
|
|
"loss": 4.3259,
|
|
"mean_token_accuracy": 0.26710256934165955,
|
|
"num_tokens": 22010927.0,
|
|
"step": 10635
|
|
},
|
|
{
|
|
"entropy": 5.25980134010315,
|
|
"epoch": 9.66424352567015,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.144544351300285e-05,
|
|
"loss": 4.3966,
|
|
"mean_token_accuracy": 0.25261815935373305,
|
|
"num_tokens": 22020033.0,
|
|
"step": 10640
|
|
},
|
|
{
|
|
"entropy": 5.301335859298706,
|
|
"epoch": 9.668786915038618,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.14057222682166e-05,
|
|
"loss": 4.5102,
|
|
"mean_token_accuracy": 0.2435757800936699,
|
|
"num_tokens": 22031124.0,
|
|
"step": 10645
|
|
},
|
|
{
|
|
"entropy": 5.258583688735962,
|
|
"epoch": 9.673330304407088,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.136655272008381e-05,
|
|
"loss": 4.4253,
|
|
"mean_token_accuracy": 0.25649013221263883,
|
|
"num_tokens": 22041624.0,
|
|
"step": 10650
|
|
},
|
|
{
|
|
"entropy": 5.271548938751221,
|
|
"epoch": 9.677873693775556,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 5.1327934965251445e-05,
|
|
"loss": 4.4851,
|
|
"mean_token_accuracy": 0.2530262812972069,
|
|
"num_tokens": 22053344.0,
|
|
"step": 10655
|
|
},
|
|
{
|
|
"entropy": 5.322299909591675,
|
|
"epoch": 9.682417083144026,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.128986909900499e-05,
|
|
"loss": 4.4186,
|
|
"mean_token_accuracy": 0.2599277004599571,
|
|
"num_tokens": 22063519.0,
|
|
"step": 10660
|
|
},
|
|
{
|
|
"entropy": 5.266521549224853,
|
|
"epoch": 9.686960472512494,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 5.125235521526814e-05,
|
|
"loss": 4.4965,
|
|
"mean_token_accuracy": 0.25153653919696806,
|
|
"num_tokens": 22074443.0,
|
|
"step": 10665
|
|
},
|
|
{
|
|
"entropy": 5.315311098098755,
|
|
"epoch": 9.691503861880964,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.121539340660269e-05,
|
|
"loss": 4.472,
|
|
"mean_token_accuracy": 0.24517737776041032,
|
|
"num_tokens": 22084664.0,
|
|
"step": 10670
|
|
},
|
|
{
|
|
"entropy": 5.282237482070923,
|
|
"epoch": 9.696047251249432,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 5.117898376420827e-05,
|
|
"loss": 4.3803,
|
|
"mean_token_accuracy": 0.261587455868721,
|
|
"num_tokens": 22093539.0,
|
|
"step": 10675
|
|
},
|
|
{
|
|
"entropy": 5.266703319549561,
|
|
"epoch": 9.7005906406179,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.1143126377922e-05,
|
|
"loss": 4.4186,
|
|
"mean_token_accuracy": 0.25147584080696106,
|
|
"num_tokens": 22103441.0,
|
|
"step": 10680
|
|
},
|
|
{
|
|
"entropy": 5.2668970108032225,
|
|
"epoch": 9.70513402998637,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 5.110782133621843e-05,
|
|
"loss": 4.4227,
|
|
"mean_token_accuracy": 0.25651430934667585,
|
|
"num_tokens": 22113770.0,
|
|
"step": 10685
|
|
},
|
|
{
|
|
"entropy": 5.242581367492676,
|
|
"epoch": 9.709677419354838,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.107306872620926e-05,
|
|
"loss": 4.452,
|
|
"mean_token_accuracy": 0.2572406202554703,
|
|
"num_tokens": 22124452.0,
|
|
"step": 10690
|
|
},
|
|
{
|
|
"entropy": 5.27496862411499,
|
|
"epoch": 9.714220808723308,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 5.10388686336431e-05,
|
|
"loss": 4.3751,
|
|
"mean_token_accuracy": 0.2537885919213295,
|
|
"num_tokens": 22133719.0,
|
|
"step": 10695
|
|
},
|
|
{
|
|
"entropy": 5.233762264251709,
|
|
"epoch": 9.718764198091776,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.1005221142905254e-05,
|
|
"loss": 4.3986,
|
|
"mean_token_accuracy": 0.2612565144896507,
|
|
"num_tokens": 22144271.0,
|
|
"step": 10700
|
|
},
|
|
{
|
|
"entropy": 5.254198741912842,
|
|
"epoch": 9.723307587460246,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.0972126337017575e-05,
|
|
"loss": 4.4449,
|
|
"mean_token_accuracy": 0.24890692979097367,
|
|
"num_tokens": 22154581.0,
|
|
"step": 10705
|
|
},
|
|
{
|
|
"entropy": 5.263562297821045,
|
|
"epoch": 9.727850976828714,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.093958429763821e-05,
|
|
"loss": 4.511,
|
|
"mean_token_accuracy": 0.24624920189380645,
|
|
"num_tokens": 22165831.0,
|
|
"step": 10710
|
|
},
|
|
{
|
|
"entropy": 5.270943117141724,
|
|
"epoch": 9.732394366197184,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 5.090759510506138e-05,
|
|
"loss": 4.4432,
|
|
"mean_token_accuracy": 0.2576790630817413,
|
|
"num_tokens": 22176407.0,
|
|
"step": 10715
|
|
},
|
|
{
|
|
"entropy": 5.25461778640747,
|
|
"epoch": 9.736937755565652,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.087615883821728e-05,
|
|
"loss": 4.4656,
|
|
"mean_token_accuracy": 0.25285989195108416,
|
|
"num_tokens": 22187451.0,
|
|
"step": 10720
|
|
},
|
|
{
|
|
"entropy": 5.28711051940918,
|
|
"epoch": 9.74148114493412,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 5.084527557467175e-05,
|
|
"loss": 4.4798,
|
|
"mean_token_accuracy": 0.24767030626535416,
|
|
"num_tokens": 22196893.0,
|
|
"step": 10725
|
|
},
|
|
{
|
|
"entropy": 5.222652864456177,
|
|
"epoch": 9.74602453430259,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.0814945390626174e-05,
|
|
"loss": 4.2926,
|
|
"mean_token_accuracy": 0.26367419958114624,
|
|
"num_tokens": 22207072.0,
|
|
"step": 10730
|
|
},
|
|
{
|
|
"entropy": 5.302778339385986,
|
|
"epoch": 9.750567923671058,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.0785168360917285e-05,
|
|
"loss": 4.4417,
|
|
"mean_token_accuracy": 0.25206770449876786,
|
|
"num_tokens": 22217708.0,
|
|
"step": 10735
|
|
},
|
|
{
|
|
"entropy": 5.23521409034729,
|
|
"epoch": 9.755111313039528,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.075594455901691e-05,
|
|
"loss": 4.414,
|
|
"mean_token_accuracy": 0.25839458256959913,
|
|
"num_tokens": 22229213.0,
|
|
"step": 10740
|
|
},
|
|
{
|
|
"entropy": 5.23668212890625,
|
|
"epoch": 9.759654702407996,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.072727405703189e-05,
|
|
"loss": 4.4266,
|
|
"mean_token_accuracy": 0.25586787015199663,
|
|
"num_tokens": 22240399.0,
|
|
"step": 10745
|
|
},
|
|
{
|
|
"entropy": 5.261511182785034,
|
|
"epoch": 9.764198091776466,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.0699156925703846e-05,
|
|
"loss": 4.4329,
|
|
"mean_token_accuracy": 0.25330416262149813,
|
|
"num_tokens": 22250842.0,
|
|
"step": 10750
|
|
},
|
|
{
|
|
"entropy": 5.25658073425293,
|
|
"epoch": 9.768741481144934,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.0671593234409e-05,
|
|
"loss": 4.3895,
|
|
"mean_token_accuracy": 0.2579048633575439,
|
|
"num_tokens": 22261829.0,
|
|
"step": 10755
|
|
},
|
|
{
|
|
"entropy": 5.286486005783081,
|
|
"epoch": 9.773284870513404,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.0644583051158005e-05,
|
|
"loss": 4.444,
|
|
"mean_token_accuracy": 0.2502687618136406,
|
|
"num_tokens": 22273515.0,
|
|
"step": 10760
|
|
},
|
|
{
|
|
"entropy": 5.22543773651123,
|
|
"epoch": 9.777828259881872,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.061812644259582e-05,
|
|
"loss": 4.4437,
|
|
"mean_token_accuracy": 0.25272453129291533,
|
|
"num_tokens": 22284167.0,
|
|
"step": 10765
|
|
},
|
|
{
|
|
"entropy": 5.231804656982422,
|
|
"epoch": 9.78237164925034,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.059222347400151e-05,
|
|
"loss": 4.4409,
|
|
"mean_token_accuracy": 0.25453758984804153,
|
|
"num_tokens": 22295442.0,
|
|
"step": 10770
|
|
},
|
|
{
|
|
"entropy": 5.316769599914551,
|
|
"epoch": 9.78691503861881,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.056687420928805e-05,
|
|
"loss": 4.4423,
|
|
"mean_token_accuracy": 0.25941198468208315,
|
|
"num_tokens": 22305717.0,
|
|
"step": 10775
|
|
},
|
|
{
|
|
"entropy": 5.25155873298645,
|
|
"epoch": 9.791458427987278,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.0542078711002246e-05,
|
|
"loss": 4.3628,
|
|
"mean_token_accuracy": 0.26655621379613875,
|
|
"num_tokens": 22315802.0,
|
|
"step": 10780
|
|
},
|
|
{
|
|
"entropy": 5.276537609100342,
|
|
"epoch": 9.796001817355748,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.0517837040324515e-05,
|
|
"loss": 4.4893,
|
|
"mean_token_accuracy": 0.23657020330429077,
|
|
"num_tokens": 22326161.0,
|
|
"step": 10785
|
|
},
|
|
{
|
|
"entropy": 5.282123136520386,
|
|
"epoch": 9.800545206724216,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.049414925706877e-05,
|
|
"loss": 4.4618,
|
|
"mean_token_accuracy": 0.24894484132528305,
|
|
"num_tokens": 22335760.0,
|
|
"step": 10790
|
|
},
|
|
{
|
|
"entropy": 5.220908069610596,
|
|
"epoch": 9.805088596092686,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.0471015419682273e-05,
|
|
"loss": 4.3834,
|
|
"mean_token_accuracy": 0.2552826151251793,
|
|
"num_tokens": 22345954.0,
|
|
"step": 10795
|
|
},
|
|
{
|
|
"entropy": 5.216185283660889,
|
|
"epoch": 9.809631985461154,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.044843558524547e-05,
|
|
"loss": 4.2911,
|
|
"mean_token_accuracy": 0.2660145416855812,
|
|
"num_tokens": 22354638.0,
|
|
"step": 10800
|
|
},
|
|
{
|
|
"entropy": 5.305073356628418,
|
|
"epoch": 9.814175374829622,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.042640980947183e-05,
|
|
"loss": 4.5073,
|
|
"mean_token_accuracy": 0.24934774935245513,
|
|
"num_tokens": 22365831.0,
|
|
"step": 10805
|
|
},
|
|
{
|
|
"entropy": 5.304891872406006,
|
|
"epoch": 9.818718764198092,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.0404938146707766e-05,
|
|
"loss": 4.459,
|
|
"mean_token_accuracy": 0.2480304554104805,
|
|
"num_tokens": 22376285.0,
|
|
"step": 10810
|
|
},
|
|
{
|
|
"entropy": 5.2086326599121096,
|
|
"epoch": 9.82326215356656,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.038402064993251e-05,
|
|
"loss": 4.4181,
|
|
"mean_token_accuracy": 0.2486609011888504,
|
|
"num_tokens": 22385587.0,
|
|
"step": 10815
|
|
},
|
|
{
|
|
"entropy": 5.254657316207886,
|
|
"epoch": 9.82780554293503,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.036365737075789e-05,
|
|
"loss": 4.4654,
|
|
"mean_token_accuracy": 0.24703366905450821,
|
|
"num_tokens": 22396273.0,
|
|
"step": 10820
|
|
},
|
|
{
|
|
"entropy": 5.254750490188599,
|
|
"epoch": 9.832348932303498,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 5.034384835942826e-05,
|
|
"loss": 4.4548,
|
|
"mean_token_accuracy": 0.2481240764260292,
|
|
"num_tokens": 22408002.0,
|
|
"step": 10825
|
|
},
|
|
{
|
|
"entropy": 5.259279775619507,
|
|
"epoch": 9.836892321671968,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.032459366482041e-05,
|
|
"loss": 4.4529,
|
|
"mean_token_accuracy": 0.24902352839708328,
|
|
"num_tokens": 22418910.0,
|
|
"step": 10830
|
|
},
|
|
{
|
|
"entropy": 5.236259460449219,
|
|
"epoch": 9.841435711040436,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.0305893334443346e-05,
|
|
"loss": 4.395,
|
|
"mean_token_accuracy": 0.2499448075890541,
|
|
"num_tokens": 22429549.0,
|
|
"step": 10835
|
|
},
|
|
{
|
|
"entropy": 5.239270830154419,
|
|
"epoch": 9.845979100408904,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.02877474144383e-05,
|
|
"loss": 4.4382,
|
|
"mean_token_accuracy": 0.24869897365570068,
|
|
"num_tokens": 22440220.0,
|
|
"step": 10840
|
|
},
|
|
{
|
|
"entropy": 5.287256336212158,
|
|
"epoch": 9.850522489777374,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.0270155949578526e-05,
|
|
"loss": 4.5063,
|
|
"mean_token_accuracy": 0.24978962242603303,
|
|
"num_tokens": 22450877.0,
|
|
"step": 10845
|
|
},
|
|
{
|
|
"entropy": 5.3022207736969,
|
|
"epoch": 9.855065879145842,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.025311898326922e-05,
|
|
"loss": 4.481,
|
|
"mean_token_accuracy": 0.251131471991539,
|
|
"num_tokens": 22461047.0,
|
|
"step": 10850
|
|
},
|
|
{
|
|
"entropy": 5.277901649475098,
|
|
"epoch": 9.859609268514312,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 5.02366365575474e-05,
|
|
"loss": 4.4665,
|
|
"mean_token_accuracy": 0.25029888451099397,
|
|
"num_tokens": 22472183.0,
|
|
"step": 10855
|
|
},
|
|
{
|
|
"entropy": 5.336268186569214,
|
|
"epoch": 9.86415265788278,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 5.022070871308179e-05,
|
|
"loss": 4.469,
|
|
"mean_token_accuracy": 0.25133287757635114,
|
|
"num_tokens": 22482576.0,
|
|
"step": 10860
|
|
},
|
|
{
|
|
"entropy": 5.254996824264526,
|
|
"epoch": 9.86869604725125,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 5.020533548917279e-05,
|
|
"loss": 4.4457,
|
|
"mean_token_accuracy": 0.2520963683724403,
|
|
"num_tokens": 22491871.0,
|
|
"step": 10865
|
|
},
|
|
{
|
|
"entropy": 5.268417310714722,
|
|
"epoch": 9.873239436619718,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 5.01905169237523e-05,
|
|
"loss": 4.4203,
|
|
"mean_token_accuracy": 0.25734980404376984,
|
|
"num_tokens": 22501738.0,
|
|
"step": 10870
|
|
},
|
|
{
|
|
"entropy": 5.21721830368042,
|
|
"epoch": 9.877782825988188,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.017625305338363e-05,
|
|
"loss": 4.3865,
|
|
"mean_token_accuracy": 0.2550504237413406,
|
|
"num_tokens": 22511692.0,
|
|
"step": 10875
|
|
},
|
|
{
|
|
"entropy": 5.278999710083008,
|
|
"epoch": 9.882326215356656,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.01625439132615e-05,
|
|
"loss": 4.4556,
|
|
"mean_token_accuracy": 0.2505263313651085,
|
|
"num_tokens": 22520975.0,
|
|
"step": 10880
|
|
},
|
|
{
|
|
"entropy": 5.319500541687011,
|
|
"epoch": 9.886869604725124,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 5.0149389537211824e-05,
|
|
"loss": 4.5024,
|
|
"mean_token_accuracy": 0.24139054268598556,
|
|
"num_tokens": 22531365.0,
|
|
"step": 10885
|
|
},
|
|
{
|
|
"entropy": 5.2468842506408695,
|
|
"epoch": 9.891412994093594,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.013678995769173e-05,
|
|
"loss": 4.4342,
|
|
"mean_token_accuracy": 0.2540967375040054,
|
|
"num_tokens": 22540949.0,
|
|
"step": 10890
|
|
},
|
|
{
|
|
"entropy": 5.255777597427368,
|
|
"epoch": 9.895956383462062,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 5.012474520578942e-05,
|
|
"loss": 4.4375,
|
|
"mean_token_accuracy": 0.2511104717850685,
|
|
"num_tokens": 22551342.0,
|
|
"step": 10895
|
|
},
|
|
{
|
|
"entropy": 5.249611330032349,
|
|
"epoch": 9.900499772830532,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 5.0113255311224107e-05,
|
|
"loss": 4.4683,
|
|
"mean_token_accuracy": 0.24963658750057222,
|
|
"num_tokens": 22562251.0,
|
|
"step": 10900
|
|
},
|
|
{
|
|
"entropy": 5.315714740753174,
|
|
"epoch": 9.905043162199,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.0102320302345996e-05,
|
|
"loss": 4.5331,
|
|
"mean_token_accuracy": 0.2435557186603546,
|
|
"num_tokens": 22573347.0,
|
|
"step": 10905
|
|
},
|
|
{
|
|
"entropy": 5.286373901367187,
|
|
"epoch": 9.90958655156747,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.0091940206136123e-05,
|
|
"loss": 4.3981,
|
|
"mean_token_accuracy": 0.26230270862579347,
|
|
"num_tokens": 22583790.0,
|
|
"step": 10910
|
|
},
|
|
{
|
|
"entropy": 5.3060774326324465,
|
|
"epoch": 9.914129940935938,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.008211504820635e-05,
|
|
"loss": 4.5238,
|
|
"mean_token_accuracy": 0.24072644710540772,
|
|
"num_tokens": 22594912.0,
|
|
"step": 10915
|
|
},
|
|
{
|
|
"entropy": 5.28284101486206,
|
|
"epoch": 9.918673330304408,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 5.007284485279925e-05,
|
|
"loss": 4.4747,
|
|
"mean_token_accuracy": 0.25395735204219816,
|
|
"num_tokens": 22605602.0,
|
|
"step": 10920
|
|
},
|
|
{
|
|
"entropy": 5.314724969863891,
|
|
"epoch": 9.923216719672876,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 5.006412964278816e-05,
|
|
"loss": 4.4651,
|
|
"mean_token_accuracy": 0.24534011781215667,
|
|
"num_tokens": 22616365.0,
|
|
"step": 10925
|
|
},
|
|
{
|
|
"entropy": 5.2651402950286865,
|
|
"epoch": 9.927760109041344,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.0055969439676925e-05,
|
|
"loss": 4.4154,
|
|
"mean_token_accuracy": 0.25878173261880877,
|
|
"num_tokens": 22626238.0,
|
|
"step": 10930
|
|
},
|
|
{
|
|
"entropy": 5.294300174713134,
|
|
"epoch": 9.932303498409814,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.004836426360008e-05,
|
|
"loss": 4.4438,
|
|
"mean_token_accuracy": 0.25455781519412995,
|
|
"num_tokens": 22636420.0,
|
|
"step": 10935
|
|
},
|
|
{
|
|
"entropy": 5.248473834991455,
|
|
"epoch": 9.936846887778282,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.004131413332263e-05,
|
|
"loss": 4.4061,
|
|
"mean_token_accuracy": 0.2586171641945839,
|
|
"num_tokens": 22646968.0,
|
|
"step": 10940
|
|
},
|
|
{
|
|
"entropy": 5.2852287769317625,
|
|
"epoch": 9.941390277146752,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.003481906624009e-05,
|
|
"loss": 4.4755,
|
|
"mean_token_accuracy": 0.247497096657753,
|
|
"num_tokens": 22657243.0,
|
|
"step": 10945
|
|
},
|
|
{
|
|
"entropy": 5.266966342926025,
|
|
"epoch": 9.94593366651522,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.002887907837835e-05,
|
|
"loss": 4.3705,
|
|
"mean_token_accuracy": 0.2659864589571953,
|
|
"num_tokens": 22668143.0,
|
|
"step": 10950
|
|
},
|
|
{
|
|
"entropy": 5.302307176589966,
|
|
"epoch": 9.95047705588369,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.002349418439379e-05,
|
|
"loss": 4.5223,
|
|
"mean_token_accuracy": 0.2425091192126274,
|
|
"num_tokens": 22680731.0,
|
|
"step": 10955
|
|
},
|
|
{
|
|
"entropy": 5.274664306640625,
|
|
"epoch": 9.955020445252158,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 5.001866439757306e-05,
|
|
"loss": 4.4428,
|
|
"mean_token_accuracy": 0.2526740238070488,
|
|
"num_tokens": 22691309.0,
|
|
"step": 10960
|
|
},
|
|
{
|
|
"entropy": 5.2655096530914305,
|
|
"epoch": 9.959563834620628,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.00143897298332e-05,
|
|
"loss": 4.4404,
|
|
"mean_token_accuracy": 0.25288567543029783,
|
|
"num_tokens": 22702636.0,
|
|
"step": 10965
|
|
},
|
|
{
|
|
"entropy": 5.260888004302979,
|
|
"epoch": 9.964107223989096,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.001067019172151e-05,
|
|
"loss": 4.4722,
|
|
"mean_token_accuracy": 0.24585205018520356,
|
|
"num_tokens": 22714801.0,
|
|
"step": 10970
|
|
},
|
|
{
|
|
"entropy": 5.260723638534546,
|
|
"epoch": 9.968650613357564,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.00075057924156e-05,
|
|
"loss": 4.4237,
|
|
"mean_token_accuracy": 0.252812060713768,
|
|
"num_tokens": 22726149.0,
|
|
"step": 10975
|
|
},
|
|
{
|
|
"entropy": 5.273374366760254,
|
|
"epoch": 9.973194002726034,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.000489653972332e-05,
|
|
"loss": 4.435,
|
|
"mean_token_accuracy": 0.25341451466083526,
|
|
"num_tokens": 22736084.0,
|
|
"step": 10980
|
|
},
|
|
{
|
|
"entropy": 5.21580753326416,
|
|
"epoch": 9.977737392094502,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.0002842440082713e-05,
|
|
"loss": 4.414,
|
|
"mean_token_accuracy": 0.256831681728363,
|
|
"num_tokens": 22745646.0,
|
|
"step": 10985
|
|
},
|
|
{
|
|
"entropy": 5.264461088180542,
|
|
"epoch": 9.982280781462972,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.000134349856208e-05,
|
|
"loss": 4.3814,
|
|
"mean_token_accuracy": 0.25695874392986295,
|
|
"num_tokens": 22755822.0,
|
|
"step": 10990
|
|
},
|
|
{
|
|
"entropy": 5.203119277954102,
|
|
"epoch": 9.98682417083144,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.000039971885989e-05,
|
|
"loss": 4.3881,
|
|
"mean_token_accuracy": 0.2644641354680061,
|
|
"num_tokens": 22765773.0,
|
|
"step": 10995
|
|
},
|
|
{
|
|
"entropy": 5.272879981994629,
|
|
"epoch": 9.99136756019991,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.0000011103304856e-05,
|
|
"loss": 4.4628,
|
|
"mean_token_accuracy": 0.2539675086736679,
|
|
"num_tokens": 22776353.0,
|
|
"step": 11000
|
|
},
|
|
{
|
|
"epoch": 9.99136756019991,
|
|
"eval_entropy": 5.1505007145016695,
|
|
"eval_loss": 5.678976535797119,
|
|
"eval_mean_token_accuracy": 0.1746614073251569,
|
|
"eval_num_tokens": 22776353.0,
|
|
"eval_runtime": 1.9956,
|
|
"eval_samples_per_second": 1721.29,
|
|
"eval_steps_per_second": 215.474,
|
|
"step": 11000
|
|
}
|
|
],
|
|
"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": true
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 4899031485143040.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|