6068 lines
164 KiB
JSON
6068 lines
164 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.725124943207633,
|
|
"eval_steps": 1000,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691572093963623,
|
|
"epoch": 0.004543389368468878,
|
|
"grad_norm": 13.0625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7962,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 10469.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.69153709411621,
|
|
"epoch": 0.009086778736937756,
|
|
"grad_norm": 12.4375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7251,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 20770.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.691234588623047,
|
|
"epoch": 0.013630168105406633,
|
|
"grad_norm": 9.625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.4762,
|
|
"mean_token_accuracy": 0.019823318906128408,
|
|
"num_tokens": 31178.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.682231140136718,
|
|
"epoch": 0.01817355747387551,
|
|
"grad_norm": 6.375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.1698,
|
|
"mean_token_accuracy": 0.04362867064774036,
|
|
"num_tokens": 41482.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.625249195098878,
|
|
"epoch": 0.02271694684234439,
|
|
"grad_norm": 3.84375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.9154,
|
|
"mean_token_accuracy": 0.04354195147752762,
|
|
"num_tokens": 52412.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.598425483703613,
|
|
"epoch": 0.027260336210813266,
|
|
"grad_norm": 3.21875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.7351,
|
|
"mean_token_accuracy": 0.042904840968549254,
|
|
"num_tokens": 62551.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.609325218200684,
|
|
"epoch": 0.03180372557928214,
|
|
"grad_norm": 3.140625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.6541,
|
|
"mean_token_accuracy": 0.042877554334700106,
|
|
"num_tokens": 72196.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.581289100646973,
|
|
"epoch": 0.03634711494775102,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.6326,
|
|
"mean_token_accuracy": 0.04586241915822029,
|
|
"num_tokens": 83372.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.568495941162109,
|
|
"epoch": 0.0408905043162199,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.6045,
|
|
"mean_token_accuracy": 0.042813677340745926,
|
|
"num_tokens": 94673.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.59875020980835,
|
|
"epoch": 0.04543389368468878,
|
|
"grad_norm": 2.703125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.4863,
|
|
"mean_token_accuracy": 0.044851933047175406,
|
|
"num_tokens": 104560.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.577375888824463,
|
|
"epoch": 0.049977283053157656,
|
|
"grad_norm": 2.515625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.471,
|
|
"mean_token_accuracy": 0.05021349526941776,
|
|
"num_tokens": 115359.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.536783027648926,
|
|
"epoch": 0.05452067242162653,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.4476,
|
|
"mean_token_accuracy": 0.046129952371120456,
|
|
"num_tokens": 126504.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.529264354705811,
|
|
"epoch": 0.059064061790095414,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.3051,
|
|
"mean_token_accuracy": 0.06012832373380661,
|
|
"num_tokens": 136449.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.412603759765625,
|
|
"epoch": 0.06360745115856428,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.155,
|
|
"mean_token_accuracy": 0.06461884267628193,
|
|
"num_tokens": 145959.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.43828363418579,
|
|
"epoch": 0.06815084052703317,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.22,
|
|
"mean_token_accuracy": 0.059075624123215674,
|
|
"num_tokens": 158533.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.466318130493164,
|
|
"epoch": 0.07269422989550205,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.0343,
|
|
"mean_token_accuracy": 0.06960411444306373,
|
|
"num_tokens": 168776.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.39024019241333,
|
|
"epoch": 0.07723761926397092,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 8.9383,
|
|
"mean_token_accuracy": 0.07182483375072479,
|
|
"num_tokens": 179528.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.281378364562988,
|
|
"epoch": 0.0817810086324398,
|
|
"grad_norm": 3.875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.8613,
|
|
"mean_token_accuracy": 0.06785103790462017,
|
|
"num_tokens": 190092.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.292144680023194,
|
|
"epoch": 0.08632439800090867,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.7131,
|
|
"mean_token_accuracy": 0.07463023066520691,
|
|
"num_tokens": 200836.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.18975601196289,
|
|
"epoch": 0.09086778736937756,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.5996,
|
|
"mean_token_accuracy": 0.07846152037382126,
|
|
"num_tokens": 210585.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.163002014160156,
|
|
"epoch": 0.09541117673784644,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.4849,
|
|
"mean_token_accuracy": 0.08028437122702599,
|
|
"num_tokens": 221439.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.084406471252441,
|
|
"epoch": 0.09995456610631531,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.4253,
|
|
"mean_token_accuracy": 0.07408776879310608,
|
|
"num_tokens": 232634.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.001629829406738,
|
|
"epoch": 0.10449795547478419,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.2732,
|
|
"mean_token_accuracy": 0.0823720321059227,
|
|
"num_tokens": 242753.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.940981578826904,
|
|
"epoch": 0.10904134484325306,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.1486,
|
|
"mean_token_accuracy": 0.08088770657777786,
|
|
"num_tokens": 252748.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.771773052215575,
|
|
"epoch": 0.11358473421172194,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.0161,
|
|
"mean_token_accuracy": 0.08380886614322662,
|
|
"num_tokens": 262772.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.701642322540284,
|
|
"epoch": 0.11812812358019083,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.9033,
|
|
"mean_token_accuracy": 0.07984891161322594,
|
|
"num_tokens": 273099.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.504860591888427,
|
|
"epoch": 0.1226715129486597,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.7777,
|
|
"mean_token_accuracy": 0.08303077444434166,
|
|
"num_tokens": 282882.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.314452743530273,
|
|
"epoch": 0.12721490231712856,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.7038,
|
|
"mean_token_accuracy": 0.08235853761434556,
|
|
"num_tokens": 294484.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.094099140167236,
|
|
"epoch": 0.13175829168559747,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.6156,
|
|
"mean_token_accuracy": 0.08289074972271919,
|
|
"num_tokens": 305154.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.792550563812256,
|
|
"epoch": 0.13630168105406634,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.4474,
|
|
"mean_token_accuracy": 0.09091320261359215,
|
|
"num_tokens": 315210.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.602887630462646,
|
|
"epoch": 0.14084507042253522,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.5223,
|
|
"mean_token_accuracy": 0.08537717014551163,
|
|
"num_tokens": 326043.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.40032892227173,
|
|
"epoch": 0.1453884597910041,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.4084,
|
|
"mean_token_accuracy": 0.0868705540895462,
|
|
"num_tokens": 335980.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.238891696929931,
|
|
"epoch": 0.14993184915947297,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.3518,
|
|
"mean_token_accuracy": 0.0846577376127243,
|
|
"num_tokens": 347094.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.232883071899414,
|
|
"epoch": 0.15447523852794184,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.3187,
|
|
"mean_token_accuracy": 0.0876984566450119,
|
|
"num_tokens": 358150.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.981767082214356,
|
|
"epoch": 0.15901862789641072,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.1827,
|
|
"mean_token_accuracy": 0.08798569217324256,
|
|
"num_tokens": 368996.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.891178703308105,
|
|
"epoch": 0.1635620172648796,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.2475,
|
|
"mean_token_accuracy": 0.09550588801503182,
|
|
"num_tokens": 378567.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.948905992507934,
|
|
"epoch": 0.16810540663334847,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.2813,
|
|
"mean_token_accuracy": 0.08694516271352767,
|
|
"num_tokens": 388461.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.791938257217407,
|
|
"epoch": 0.17264879600181735,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.1569,
|
|
"mean_token_accuracy": 0.09532128497958184,
|
|
"num_tokens": 399353.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.853602457046509,
|
|
"epoch": 0.17719218537028622,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.2436,
|
|
"mean_token_accuracy": 0.08666228726506234,
|
|
"num_tokens": 409282.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.708158397674561,
|
|
"epoch": 0.18173557473875512,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.1465,
|
|
"mean_token_accuracy": 0.09066225066781045,
|
|
"num_tokens": 420844.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.8060750484466555,
|
|
"epoch": 0.186278964107224,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.1545,
|
|
"mean_token_accuracy": 0.08872071430087089,
|
|
"num_tokens": 431681.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.521131896972657,
|
|
"epoch": 0.19082235347569287,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.1005,
|
|
"mean_token_accuracy": 0.09334186390042305,
|
|
"num_tokens": 441617.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.723299789428711,
|
|
"epoch": 0.19536574284416175,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.0738,
|
|
"mean_token_accuracy": 0.09614492207765579,
|
|
"num_tokens": 451165.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.528104972839356,
|
|
"epoch": 0.19990913221263062,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.047,
|
|
"mean_token_accuracy": 0.09558017253875732,
|
|
"num_tokens": 461205.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.59382004737854,
|
|
"epoch": 0.2044525215810995,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.0261,
|
|
"mean_token_accuracy": 0.09080673456192016,
|
|
"num_tokens": 471409.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.570555877685547,
|
|
"epoch": 0.20899591094956838,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.0239,
|
|
"mean_token_accuracy": 0.09434545040130615,
|
|
"num_tokens": 480741.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.4990636825561525,
|
|
"epoch": 0.21353930031803725,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.0382,
|
|
"mean_token_accuracy": 0.09171517789363862,
|
|
"num_tokens": 491576.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.489644718170166,
|
|
"epoch": 0.21808268968650613,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.0266,
|
|
"mean_token_accuracy": 0.09193908870220184,
|
|
"num_tokens": 503121.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.529499769210815,
|
|
"epoch": 0.222626079054975,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.9917,
|
|
"mean_token_accuracy": 0.08934131860733033,
|
|
"num_tokens": 513956.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.528982019424438,
|
|
"epoch": 0.22716946842344388,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.0225,
|
|
"mean_token_accuracy": 0.09366922527551651,
|
|
"num_tokens": 525043.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.4644406795501705,
|
|
"epoch": 0.23171285779191278,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.9558,
|
|
"mean_token_accuracy": 0.10055001303553582,
|
|
"num_tokens": 535824.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.484562730789184,
|
|
"epoch": 0.23625624716038165,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.9765,
|
|
"mean_token_accuracy": 0.09478841349482536,
|
|
"num_tokens": 546553.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.449831628799439,
|
|
"epoch": 0.24079963652885053,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.9221,
|
|
"mean_token_accuracy": 0.09788105860352517,
|
|
"num_tokens": 556637.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.346249198913574,
|
|
"epoch": 0.2453430258973194,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.9499,
|
|
"mean_token_accuracy": 0.09590627774596214,
|
|
"num_tokens": 568464.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.440023136138916,
|
|
"epoch": 0.24988641526578828,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.9988,
|
|
"mean_token_accuracy": 0.09769970029592515,
|
|
"num_tokens": 579387.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.344238996505737,
|
|
"epoch": 0.25442980463425713,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.0771,
|
|
"mean_token_accuracy": 0.09503750577569008,
|
|
"num_tokens": 590524.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.465632104873658,
|
|
"epoch": 0.25897319400272606,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.9365,
|
|
"mean_token_accuracy": 0.09854642376303673,
|
|
"num_tokens": 601570.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.3490095138549805,
|
|
"epoch": 0.26351658337119493,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.8802,
|
|
"mean_token_accuracy": 0.09717997089028359,
|
|
"num_tokens": 611615.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.340326023101807,
|
|
"epoch": 0.2680599727396638,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.9172,
|
|
"mean_token_accuracy": 0.09668021574616432,
|
|
"num_tokens": 621077.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.310168886184693,
|
|
"epoch": 0.2726033621081327,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.865,
|
|
"mean_token_accuracy": 0.09947703778743744,
|
|
"num_tokens": 631235.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.346650838851929,
|
|
"epoch": 0.27714675147660156,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.8915,
|
|
"mean_token_accuracy": 0.10305262729525566,
|
|
"num_tokens": 641195.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.243460988998413,
|
|
"epoch": 0.28169014084507044,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.899,
|
|
"mean_token_accuracy": 0.09924287423491478,
|
|
"num_tokens": 651420.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.327885532379151,
|
|
"epoch": 0.2862335302135393,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.8654,
|
|
"mean_token_accuracy": 0.10240009650588036,
|
|
"num_tokens": 661370.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.278406858444214,
|
|
"epoch": 0.2907769195820082,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.8022,
|
|
"mean_token_accuracy": 0.09714617282152176,
|
|
"num_tokens": 672237.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.201946496963501,
|
|
"epoch": 0.29532030895047706,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.8708,
|
|
"mean_token_accuracy": 0.1012646809220314,
|
|
"num_tokens": 682478.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.349485778808594,
|
|
"epoch": 0.29986369831894594,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.9835,
|
|
"mean_token_accuracy": 0.09188964888453484,
|
|
"num_tokens": 693669.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.099835443496704,
|
|
"epoch": 0.3044070876874148,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.7072,
|
|
"mean_token_accuracy": 0.10328638926148415,
|
|
"num_tokens": 704260.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.186357593536377,
|
|
"epoch": 0.3089504770558837,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.6935,
|
|
"mean_token_accuracy": 0.1045442394912243,
|
|
"num_tokens": 713924.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.213700151443481,
|
|
"epoch": 0.31349386642435256,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.8839,
|
|
"mean_token_accuracy": 0.0973054438829422,
|
|
"num_tokens": 724208.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.211179637908936,
|
|
"epoch": 0.31803725579282144,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.7871,
|
|
"mean_token_accuracy": 0.10154509842395783,
|
|
"num_tokens": 733730.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.0504943370819095,
|
|
"epoch": 0.3225806451612903,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.745,
|
|
"mean_token_accuracy": 0.10221930891275406,
|
|
"num_tokens": 743467.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.2732799530029295,
|
|
"epoch": 0.3271240345297592,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.878,
|
|
"mean_token_accuracy": 0.10214311555027962,
|
|
"num_tokens": 753446.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.161293411254883,
|
|
"epoch": 0.33166742389822806,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.8045,
|
|
"mean_token_accuracy": 0.09740775153040886,
|
|
"num_tokens": 763257.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.1125328540802,
|
|
"epoch": 0.33621081326669694,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.7808,
|
|
"mean_token_accuracy": 0.10515827611088753,
|
|
"num_tokens": 773322.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.063180541992187,
|
|
"epoch": 0.3407542026351658,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.8664,
|
|
"mean_token_accuracy": 0.09914991706609726,
|
|
"num_tokens": 784297.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.194390058517456,
|
|
"epoch": 0.3452975920036347,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.8432,
|
|
"mean_token_accuracy": 0.0979395940899849,
|
|
"num_tokens": 795186.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.138232707977295,
|
|
"epoch": 0.34984098137210357,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.7456,
|
|
"mean_token_accuracy": 0.10714830458164215,
|
|
"num_tokens": 805802.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.068744134902954,
|
|
"epoch": 0.35438437074057244,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.7036,
|
|
"mean_token_accuracy": 0.10414793938398362,
|
|
"num_tokens": 816581.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.050993919372559,
|
|
"epoch": 0.35892776010904137,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.7162,
|
|
"mean_token_accuracy": 0.10066760554909707,
|
|
"num_tokens": 826724.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.132152462005616,
|
|
"epoch": 0.36347114947751025,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.7557,
|
|
"mean_token_accuracy": 0.10789255201816558,
|
|
"num_tokens": 836431.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 6.924490213394165,
|
|
"epoch": 0.3680145388459791,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.673,
|
|
"mean_token_accuracy": 0.10689961239695549,
|
|
"num_tokens": 846888.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.120433712005616,
|
|
"epoch": 0.372557928214448,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.7498,
|
|
"mean_token_accuracy": 0.10389859899878502,
|
|
"num_tokens": 856062.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.0627799987792965,
|
|
"epoch": 0.3771013175829169,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.7656,
|
|
"mean_token_accuracy": 0.10278895348310471,
|
|
"num_tokens": 866489.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 6.958880233764648,
|
|
"epoch": 0.38164470695138575,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.6184,
|
|
"mean_token_accuracy": 0.1000476323068142,
|
|
"num_tokens": 876895.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.133209085464477,
|
|
"epoch": 0.3861880963198546,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.7897,
|
|
"mean_token_accuracy": 0.10413167849183083,
|
|
"num_tokens": 886931.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.022590970993042,
|
|
"epoch": 0.3907314856883235,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.7075,
|
|
"mean_token_accuracy": 0.10023215562105178,
|
|
"num_tokens": 896229.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 6.995979881286621,
|
|
"epoch": 0.3952748750567924,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.6555,
|
|
"mean_token_accuracy": 0.10956079438328743,
|
|
"num_tokens": 906668.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.940561723709107,
|
|
"epoch": 0.39981826442526125,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.648,
|
|
"mean_token_accuracy": 0.10910732224583626,
|
|
"num_tokens": 915868.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 6.948301935195923,
|
|
"epoch": 0.4043616537937301,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.7005,
|
|
"mean_token_accuracy": 0.10821465030312538,
|
|
"num_tokens": 926050.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.959620666503906,
|
|
"epoch": 0.408905043162199,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.5741,
|
|
"mean_token_accuracy": 0.1120890200138092,
|
|
"num_tokens": 936543.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.945019960403442,
|
|
"epoch": 0.4134484325306679,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.7137,
|
|
"mean_token_accuracy": 0.0983024187386036,
|
|
"num_tokens": 946895.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.044814252853394,
|
|
"epoch": 0.41799182189913675,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.7416,
|
|
"mean_token_accuracy": 0.10709839686751366,
|
|
"num_tokens": 956074.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.954236841201782,
|
|
"epoch": 0.4225352112676056,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.6696,
|
|
"mean_token_accuracy": 0.10691480711102486,
|
|
"num_tokens": 966264.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.965500164031982,
|
|
"epoch": 0.4270786006360745,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.7345,
|
|
"mean_token_accuracy": 0.1096844032406807,
|
|
"num_tokens": 976369.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.890221691131591,
|
|
"epoch": 0.4316219900045434,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.7048,
|
|
"mean_token_accuracy": 0.11195428147912026,
|
|
"num_tokens": 986948.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.929348278045654,
|
|
"epoch": 0.43616537937301225,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.6449,
|
|
"mean_token_accuracy": 0.10682991668581962,
|
|
"num_tokens": 996920.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.977197074890137,
|
|
"epoch": 0.4407087687414811,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.6733,
|
|
"mean_token_accuracy": 0.10687830671668053,
|
|
"num_tokens": 1006343.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.894662284851075,
|
|
"epoch": 0.44525215810995,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.5875,
|
|
"mean_token_accuracy": 0.1086734913289547,
|
|
"num_tokens": 1017156.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.851984214782715,
|
|
"epoch": 0.4497955474784189,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.6272,
|
|
"mean_token_accuracy": 0.10609063655138015,
|
|
"num_tokens": 1027082.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.872963380813599,
|
|
"epoch": 0.45433893684688775,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.6524,
|
|
"mean_token_accuracy": 0.10921891406178474,
|
|
"num_tokens": 1037458.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.879157161712646,
|
|
"epoch": 0.45888232621535663,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.6272,
|
|
"mean_token_accuracy": 0.10627908930182457,
|
|
"num_tokens": 1048391.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.818037796020508,
|
|
"epoch": 0.46342571558382556,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.5941,
|
|
"mean_token_accuracy": 0.10977126732468605,
|
|
"num_tokens": 1058281.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.928041458129883,
|
|
"epoch": 0.46796910495229443,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.6453,
|
|
"mean_token_accuracy": 0.10617673844099045,
|
|
"num_tokens": 1069157.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.9156416893005375,
|
|
"epoch": 0.4725124943207633,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.6717,
|
|
"mean_token_accuracy": 0.11312429681420326,
|
|
"num_tokens": 1079986.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.849049139022827,
|
|
"epoch": 0.4770558836892322,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.6691,
|
|
"mean_token_accuracy": 0.1143962450325489,
|
|
"num_tokens": 1090414.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.789116430282593,
|
|
"epoch": 0.48159927305770106,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.5645,
|
|
"mean_token_accuracy": 0.11084540784358979,
|
|
"num_tokens": 1100909.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.932785320281982,
|
|
"epoch": 0.48614266242616994,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6774,
|
|
"mean_token_accuracy": 0.11115043088793755,
|
|
"num_tokens": 1110780.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.846625757217407,
|
|
"epoch": 0.4906860517946388,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.5804,
|
|
"mean_token_accuracy": 0.10537252500653267,
|
|
"num_tokens": 1121669.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.735438680648803,
|
|
"epoch": 0.4952294411631077,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.5973,
|
|
"mean_token_accuracy": 0.11126141846179963,
|
|
"num_tokens": 1131892.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.797910976409912,
|
|
"epoch": 0.49977283053157656,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.5769,
|
|
"mean_token_accuracy": 0.10420559868216514,
|
|
"num_tokens": 1143911.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.850543737411499,
|
|
"epoch": 0.5043162199000454,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.6045,
|
|
"mean_token_accuracy": 0.10880815088748932,
|
|
"num_tokens": 1154826.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.842674684524536,
|
|
"epoch": 0.5088596092685143,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.5888,
|
|
"mean_token_accuracy": 0.1078294150531292,
|
|
"num_tokens": 1164343.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.748653364181519,
|
|
"epoch": 0.5134029986369832,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.6014,
|
|
"mean_token_accuracy": 0.11128783077001572,
|
|
"num_tokens": 1174447.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.780460023880005,
|
|
"epoch": 0.5179463880054521,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.4939,
|
|
"mean_token_accuracy": 0.10823575183749198,
|
|
"num_tokens": 1186092.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.844918966293335,
|
|
"epoch": 0.522489777373921,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.6552,
|
|
"mean_token_accuracy": 0.11681408137083053,
|
|
"num_tokens": 1195151.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.7070849418640135,
|
|
"epoch": 0.5270331667423899,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.4827,
|
|
"mean_token_accuracy": 0.11342033669352532,
|
|
"num_tokens": 1204756.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.668478345870971,
|
|
"epoch": 0.5315765561108587,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.4956,
|
|
"mean_token_accuracy": 0.11778365448117256,
|
|
"num_tokens": 1213573.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.8163121223449705,
|
|
"epoch": 0.5361199454793276,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.5003,
|
|
"mean_token_accuracy": 0.10915512219071388,
|
|
"num_tokens": 1224031.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.67197813987732,
|
|
"epoch": 0.5406633348477965,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.5227,
|
|
"mean_token_accuracy": 0.11702188551425934,
|
|
"num_tokens": 1233679.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.769627475738526,
|
|
"epoch": 0.5452067242162654,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.5126,
|
|
"mean_token_accuracy": 0.12154011651873589,
|
|
"num_tokens": 1243877.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.721151685714721,
|
|
"epoch": 0.5497501135847342,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.496,
|
|
"mean_token_accuracy": 0.12043491527438163,
|
|
"num_tokens": 1253913.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.748045492172241,
|
|
"epoch": 0.5542935029532031,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.5398,
|
|
"mean_token_accuracy": 0.11470362469553948,
|
|
"num_tokens": 1263895.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.681035137176513,
|
|
"epoch": 0.558836892321672,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.5484,
|
|
"mean_token_accuracy": 0.10923809856176377,
|
|
"num_tokens": 1273695.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.7565836906433105,
|
|
"epoch": 0.5633802816901409,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.6568,
|
|
"mean_token_accuracy": 0.11423956826329232,
|
|
"num_tokens": 1285093.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.78377594947815,
|
|
"epoch": 0.5679236710586097,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.6084,
|
|
"mean_token_accuracy": 0.1104039043188095,
|
|
"num_tokens": 1295354.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.641982460021973,
|
|
"epoch": 0.5724670604270786,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.5779,
|
|
"mean_token_accuracy": 0.10812898948788643,
|
|
"num_tokens": 1305869.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.779891061782837,
|
|
"epoch": 0.5770104497955475,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.5507,
|
|
"mean_token_accuracy": 0.11241080686450004,
|
|
"num_tokens": 1315845.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.662394618988037,
|
|
"epoch": 0.5815538391640164,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.5999,
|
|
"mean_token_accuracy": 0.11409554556012154,
|
|
"num_tokens": 1326056.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.768114376068115,
|
|
"epoch": 0.5860972285324852,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.5209,
|
|
"mean_token_accuracy": 0.11618900150060654,
|
|
"num_tokens": 1336843.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.638251924514771,
|
|
"epoch": 0.5906406179009541,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.3885,
|
|
"mean_token_accuracy": 0.12263384908437729,
|
|
"num_tokens": 1346106.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.7041887760162355,
|
|
"epoch": 0.595184007269423,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.498,
|
|
"mean_token_accuracy": 0.11787696108222008,
|
|
"num_tokens": 1356321.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.629571962356567,
|
|
"epoch": 0.5997273966378919,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.5542,
|
|
"mean_token_accuracy": 0.11547330170869827,
|
|
"num_tokens": 1367452.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.644501113891602,
|
|
"epoch": 0.6042707860063607,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.4496,
|
|
"mean_token_accuracy": 0.12300742790102959,
|
|
"num_tokens": 1377099.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.647049760818481,
|
|
"epoch": 0.6088141753748296,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.4602,
|
|
"mean_token_accuracy": 0.1089586578309536,
|
|
"num_tokens": 1387768.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.568883752822876,
|
|
"epoch": 0.6133575647432985,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.4266,
|
|
"mean_token_accuracy": 0.11893959119915962,
|
|
"num_tokens": 1398703.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.646634721755982,
|
|
"epoch": 0.6179009541117674,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.4899,
|
|
"mean_token_accuracy": 0.11673004627227783,
|
|
"num_tokens": 1409059.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.548872947692871,
|
|
"epoch": 0.6224443434802363,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.4303,
|
|
"mean_token_accuracy": 0.12090117484331131,
|
|
"num_tokens": 1420100.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.630219030380249,
|
|
"epoch": 0.6269877328487051,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.4593,
|
|
"mean_token_accuracy": 0.11943159177899361,
|
|
"num_tokens": 1430052.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.608811616897583,
|
|
"epoch": 0.631531122217174,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.5102,
|
|
"mean_token_accuracy": 0.11295234113931656,
|
|
"num_tokens": 1441466.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.656955623626709,
|
|
"epoch": 0.6360745115856429,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.4618,
|
|
"mean_token_accuracy": 0.11585693284869195,
|
|
"num_tokens": 1451839.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.55370454788208,
|
|
"epoch": 0.6406179009541118,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.4039,
|
|
"mean_token_accuracy": 0.11986656337976456,
|
|
"num_tokens": 1461287.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.56372971534729,
|
|
"epoch": 0.6451612903225806,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.5055,
|
|
"mean_token_accuracy": 0.10829323157668114,
|
|
"num_tokens": 1471912.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.684666442871094,
|
|
"epoch": 0.6497046796910495,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.5568,
|
|
"mean_token_accuracy": 0.11624274030327797,
|
|
"num_tokens": 1482912.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.4966638565063475,
|
|
"epoch": 0.6542480690595184,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4357,
|
|
"mean_token_accuracy": 0.11463170126080513,
|
|
"num_tokens": 1493226.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.621252202987671,
|
|
"epoch": 0.6587914584279873,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.5498,
|
|
"mean_token_accuracy": 0.11490358039736748,
|
|
"num_tokens": 1503411.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.703120422363281,
|
|
"epoch": 0.6633348477964561,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.5453,
|
|
"mean_token_accuracy": 0.11521322354674339,
|
|
"num_tokens": 1513715.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.6506260395050045,
|
|
"epoch": 0.667878237164925,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.5034,
|
|
"mean_token_accuracy": 0.11211693659424782,
|
|
"num_tokens": 1524309.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.5112879276275635,
|
|
"epoch": 0.6724216265333939,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5256,
|
|
"mean_token_accuracy": 0.1152856782078743,
|
|
"num_tokens": 1534901.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.618919229507446,
|
|
"epoch": 0.6769650159018628,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.3946,
|
|
"mean_token_accuracy": 0.11816890388727189,
|
|
"num_tokens": 1544576.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.500910997390747,
|
|
"epoch": 0.6815084052703316,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.4213,
|
|
"mean_token_accuracy": 0.11983850598335266,
|
|
"num_tokens": 1554800.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.617109727859497,
|
|
"epoch": 0.6860517946388005,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.4735,
|
|
"mean_token_accuracy": 0.12086164802312852,
|
|
"num_tokens": 1565166.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.552991437911987,
|
|
"epoch": 0.6905951840072694,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.4959,
|
|
"mean_token_accuracy": 0.1172470510005951,
|
|
"num_tokens": 1576211.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.579512023925782,
|
|
"epoch": 0.6951385733757383,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.4105,
|
|
"mean_token_accuracy": 0.11526432931423188,
|
|
"num_tokens": 1586444.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.611948537826538,
|
|
"epoch": 0.6996819627442071,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.4112,
|
|
"mean_token_accuracy": 0.12121847867965699,
|
|
"num_tokens": 1596886.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.387885427474975,
|
|
"epoch": 0.704225352112676,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.3065,
|
|
"mean_token_accuracy": 0.12513158097863197,
|
|
"num_tokens": 1606383.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.478776264190674,
|
|
"epoch": 0.7087687414811449,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.4026,
|
|
"mean_token_accuracy": 0.11760507896542549,
|
|
"num_tokens": 1616747.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.545230388641357,
|
|
"epoch": 0.7133121308496138,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.3592,
|
|
"mean_token_accuracy": 0.12122382372617721,
|
|
"num_tokens": 1626760.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.550805711746216,
|
|
"epoch": 0.7178555202180827,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.5778,
|
|
"mean_token_accuracy": 0.11731743440032005,
|
|
"num_tokens": 1636419.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.451327610015869,
|
|
"epoch": 0.7223989095865516,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.3539,
|
|
"mean_token_accuracy": 0.12045771181583405,
|
|
"num_tokens": 1646349.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.549941158294677,
|
|
"epoch": 0.7269422989550205,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.425,
|
|
"mean_token_accuracy": 0.12031189724802971,
|
|
"num_tokens": 1657693.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.502531576156616,
|
|
"epoch": 0.7314856883234894,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.369,
|
|
"mean_token_accuracy": 0.11813340857625007,
|
|
"num_tokens": 1667689.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.473275232315063,
|
|
"epoch": 0.7360290776919582,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.3412,
|
|
"mean_token_accuracy": 0.127835476398468,
|
|
"num_tokens": 1677402.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.365635585784912,
|
|
"epoch": 0.7405724670604271,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.2735,
|
|
"mean_token_accuracy": 0.12139133512973785,
|
|
"num_tokens": 1687285.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.558733177185059,
|
|
"epoch": 0.745115856428896,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.5708,
|
|
"mean_token_accuracy": 0.10980626493692398,
|
|
"num_tokens": 1697784.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.504992198944092,
|
|
"epoch": 0.7496592457973649,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.3909,
|
|
"mean_token_accuracy": 0.11993959099054337,
|
|
"num_tokens": 1708313.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.4220723628997805,
|
|
"epoch": 0.7542026351658337,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.3418,
|
|
"mean_token_accuracy": 0.12350277826189995,
|
|
"num_tokens": 1718278.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.451784753799439,
|
|
"epoch": 0.7587460245343026,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.3756,
|
|
"mean_token_accuracy": 0.119990985840559,
|
|
"num_tokens": 1729430.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.453255462646484,
|
|
"epoch": 0.7632894139027715,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.3419,
|
|
"mean_token_accuracy": 0.12214131876826287,
|
|
"num_tokens": 1740759.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.459233808517456,
|
|
"epoch": 0.7678328032712404,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.4063,
|
|
"mean_token_accuracy": 0.1200601689517498,
|
|
"num_tokens": 1751519.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.4423778533935545,
|
|
"epoch": 0.7723761926397092,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.3652,
|
|
"mean_token_accuracy": 0.12372113093733787,
|
|
"num_tokens": 1761731.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.378735780715942,
|
|
"epoch": 0.7769195820081781,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.2778,
|
|
"mean_token_accuracy": 0.12224024012684823,
|
|
"num_tokens": 1771949.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.358287000656128,
|
|
"epoch": 0.781462971376647,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.3825,
|
|
"mean_token_accuracy": 0.12915819510817528,
|
|
"num_tokens": 1782923.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.5328343391418455,
|
|
"epoch": 0.7860063607451159,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.4885,
|
|
"mean_token_accuracy": 0.1134648710489273,
|
|
"num_tokens": 1792986.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.601466178894043,
|
|
"epoch": 0.7905497501135847,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.4821,
|
|
"mean_token_accuracy": 0.11608704775571824,
|
|
"num_tokens": 1804820.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.368373203277588,
|
|
"epoch": 0.7950931394820536,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.4323,
|
|
"mean_token_accuracy": 0.12307631820440293,
|
|
"num_tokens": 1815360.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.398289680480957,
|
|
"epoch": 0.7996365288505225,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.4034,
|
|
"mean_token_accuracy": 0.11827710643410683,
|
|
"num_tokens": 1826374.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.460321140289307,
|
|
"epoch": 0.8041799182189914,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.3476,
|
|
"mean_token_accuracy": 0.11750650107860565,
|
|
"num_tokens": 1836535.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.503727102279663,
|
|
"epoch": 0.8087233075874602,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.4294,
|
|
"mean_token_accuracy": 0.11657789722084999,
|
|
"num_tokens": 1846678.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.358234071731568,
|
|
"epoch": 0.8132666969559291,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.4212,
|
|
"mean_token_accuracy": 0.11958692297339439,
|
|
"num_tokens": 1856883.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.430210161209106,
|
|
"epoch": 0.817810086324398,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.3803,
|
|
"mean_token_accuracy": 0.12223695889115334,
|
|
"num_tokens": 1866814.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.4646649837493895,
|
|
"epoch": 0.8223534756928669,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.4072,
|
|
"mean_token_accuracy": 0.12242427542805671,
|
|
"num_tokens": 1877136.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.434080982208252,
|
|
"epoch": 0.8268968650613358,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.4273,
|
|
"mean_token_accuracy": 0.12286214902997017,
|
|
"num_tokens": 1887801.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.391744661331177,
|
|
"epoch": 0.8314402544298046,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.243,
|
|
"mean_token_accuracy": 0.12523403987288476,
|
|
"num_tokens": 1898313.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.383115625381469,
|
|
"epoch": 0.8359836437982735,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.3005,
|
|
"mean_token_accuracy": 0.1225958451628685,
|
|
"num_tokens": 1909306.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.445513391494751,
|
|
"epoch": 0.8405270331667424,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.3765,
|
|
"mean_token_accuracy": 0.12136013731360436,
|
|
"num_tokens": 1919406.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.3957679271698,
|
|
"epoch": 0.8450704225352113,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.2993,
|
|
"mean_token_accuracy": 0.12907271087169647,
|
|
"num_tokens": 1929087.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.272099876403809,
|
|
"epoch": 0.8496138119036801,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.2391,
|
|
"mean_token_accuracy": 0.12513313740491866,
|
|
"num_tokens": 1938647.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.347430801391601,
|
|
"epoch": 0.854157201272149,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.2817,
|
|
"mean_token_accuracy": 0.12321201413869858,
|
|
"num_tokens": 1948237.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.499377059936523,
|
|
"epoch": 0.8587005906406179,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.3808,
|
|
"mean_token_accuracy": 0.12037008255720139,
|
|
"num_tokens": 1958440.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.301383686065674,
|
|
"epoch": 0.8632439800090868,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.2939,
|
|
"mean_token_accuracy": 0.12816702499985694,
|
|
"num_tokens": 1968388.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.333843803405761,
|
|
"epoch": 0.8677873693775556,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.3915,
|
|
"mean_token_accuracy": 0.11608542948961258,
|
|
"num_tokens": 1979479.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.333086395263672,
|
|
"epoch": 0.8723307587460245,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.2446,
|
|
"mean_token_accuracy": 0.12511033266782762,
|
|
"num_tokens": 1989680.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.302000331878662,
|
|
"epoch": 0.8768741481144934,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.2962,
|
|
"mean_token_accuracy": 0.12111640349030495,
|
|
"num_tokens": 2000489.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.3346357345581055,
|
|
"epoch": 0.8814175374829623,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.2418,
|
|
"mean_token_accuracy": 0.12388192564249038,
|
|
"num_tokens": 2010299.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.205083179473877,
|
|
"epoch": 0.8859609268514311,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.2126,
|
|
"mean_token_accuracy": 0.1265421062707901,
|
|
"num_tokens": 2019374.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.314967250823974,
|
|
"epoch": 0.8905043162199,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.2074,
|
|
"mean_token_accuracy": 0.13389402404427528,
|
|
"num_tokens": 2029222.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.326553821563721,
|
|
"epoch": 0.8950477055883689,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.3335,
|
|
"mean_token_accuracy": 0.12566622868180274,
|
|
"num_tokens": 2041425.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.355471420288086,
|
|
"epoch": 0.8995910949568378,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.4146,
|
|
"mean_token_accuracy": 0.12330245301127433,
|
|
"num_tokens": 2052618.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.292330026626587,
|
|
"epoch": 0.9041344843253066,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.2855,
|
|
"mean_token_accuracy": 0.12251581624150276,
|
|
"num_tokens": 2064059.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.2780286312103275,
|
|
"epoch": 0.9086778736937755,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.3067,
|
|
"mean_token_accuracy": 0.12127709686756134,
|
|
"num_tokens": 2075341.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.9086778736937755,
|
|
"eval_entropy": 6.218727421206097,
|
|
"eval_loss": 6.430349349975586,
|
|
"eval_mean_token_accuracy": 0.12361673391315826,
|
|
"eval_num_tokens": 2075341.0,
|
|
"eval_runtime": 2.6331,
|
|
"eval_samples_per_second": 1304.534,
|
|
"eval_steps_per_second": 163.304,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.206767177581787,
|
|
"epoch": 0.9132212630622444,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999998223471442,
|
|
"loss": 6.1819,
|
|
"mean_token_accuracy": 0.13016388192772865,
|
|
"num_tokens": 2085683.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.341091871261597,
|
|
"epoch": 0.9177646524307133,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999991006328981,
|
|
"loss": 6.3006,
|
|
"mean_token_accuracy": 0.12977074533700944,
|
|
"num_tokens": 2095183.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.186359834671021,
|
|
"epoch": 0.9223080417991822,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999978237557377,
|
|
"loss": 6.167,
|
|
"mean_token_accuracy": 0.13407007083296776,
|
|
"num_tokens": 2104948.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.191218757629395,
|
|
"epoch": 0.9268514311676511,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999959917188137,
|
|
"loss": 6.1496,
|
|
"mean_token_accuracy": 0.13299431800842285,
|
|
"num_tokens": 2115122.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.343822002410889,
|
|
"epoch": 0.93139482053612,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999936045266462,
|
|
"loss": 6.2475,
|
|
"mean_token_accuracy": 0.12717320919036865,
|
|
"num_tokens": 2125665.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.183660650253296,
|
|
"epoch": 0.9359382099045889,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999906621851255,
|
|
"loss": 6.1877,
|
|
"mean_token_accuracy": 0.1269860491156578,
|
|
"num_tokens": 2137119.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.218777894973755,
|
|
"epoch": 0.9404815992730577,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999871647015115,
|
|
"loss": 6.2307,
|
|
"mean_token_accuracy": 0.12949277758598327,
|
|
"num_tokens": 2147506.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.268872499465942,
|
|
"epoch": 0.9450249886415266,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999831120844339,
|
|
"loss": 6.2461,
|
|
"mean_token_accuracy": 0.13204365074634553,
|
|
"num_tokens": 2157054.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.353840112686157,
|
|
"epoch": 0.9495683780099955,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999785043438922,
|
|
"loss": 6.3052,
|
|
"mean_token_accuracy": 0.12659802213311194,
|
|
"num_tokens": 2167333.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.18822603225708,
|
|
"epoch": 0.9541117673784644,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999733414912555,
|
|
"loss": 6.2603,
|
|
"mean_token_accuracy": 0.12484889551997184,
|
|
"num_tokens": 2177876.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.28578929901123,
|
|
"epoch": 0.9586551567469332,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999676235392626,
|
|
"loss": 6.1924,
|
|
"mean_token_accuracy": 0.13271888494491577,
|
|
"num_tokens": 2187184.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.275188207626343,
|
|
"epoch": 0.9631985461154021,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999613505020219,
|
|
"loss": 6.23,
|
|
"mean_token_accuracy": 0.12417279556393623,
|
|
"num_tokens": 2197258.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.218803834915161,
|
|
"epoch": 0.967741935483871,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999545223950116,
|
|
"loss": 6.2842,
|
|
"mean_token_accuracy": 0.1329876147210598,
|
|
"num_tokens": 2207440.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.341020202636718,
|
|
"epoch": 0.9722853248523399,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999471392350795,
|
|
"loss": 6.3201,
|
|
"mean_token_accuracy": 0.13476093709468842,
|
|
"num_tokens": 2216454.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.273011779785156,
|
|
"epoch": 0.9768287142208087,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004999392010404426,
|
|
"loss": 6.1687,
|
|
"mean_token_accuracy": 0.12653264328837394,
|
|
"num_tokens": 2226270.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.155029344558716,
|
|
"epoch": 0.9813721035892776,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999307078306876,
|
|
"loss": 6.2223,
|
|
"mean_token_accuracy": 0.12701170668005943,
|
|
"num_tokens": 2235698.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.341498994827271,
|
|
"epoch": 0.9859154929577465,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999216596267708,
|
|
"loss": 6.2563,
|
|
"mean_token_accuracy": 0.13341840505599975,
|
|
"num_tokens": 2246309.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.215201282501221,
|
|
"epoch": 0.9904588823262154,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999120564510177,
|
|
"loss": 6.2721,
|
|
"mean_token_accuracy": 0.13170336931943893,
|
|
"num_tokens": 2257258.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.1683765888214115,
|
|
"epoch": 0.9950022716946842,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499901898327123,
|
|
"loss": 6.2068,
|
|
"mean_token_accuracy": 0.12499793097376824,
|
|
"num_tokens": 2267825.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.333638763427734,
|
|
"epoch": 0.9995456610631531,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998911852801512,
|
|
"loss": 6.3206,
|
|
"mean_token_accuracy": 0.12815747186541557,
|
|
"num_tokens": 2278727.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.311305787828234,
|
|
"epoch": 1.003634711494775,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998799173365355,
|
|
"loss": 6.1588,
|
|
"mean_token_accuracy": 0.1277164104912016,
|
|
"num_tokens": 2287306.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.1962635040283205,
|
|
"epoch": 1.008178100863244,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998680945240783,
|
|
"loss": 6.1096,
|
|
"mean_token_accuracy": 0.13211133405566217,
|
|
"num_tokens": 2299208.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.3119384765625,
|
|
"epoch": 1.0127214902317128,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998557168719513,
|
|
"loss": 6.06,
|
|
"mean_token_accuracy": 0.13208466172218322,
|
|
"num_tokens": 2309196.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.21234393119812,
|
|
"epoch": 1.0172648796001817,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998427844106954,
|
|
"loss": 6.1639,
|
|
"mean_token_accuracy": 0.12953464984893798,
|
|
"num_tokens": 2320255.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.236302614212036,
|
|
"epoch": 1.0218082689686505,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004998292971722197,
|
|
"loss": 6.0808,
|
|
"mean_token_accuracy": 0.1310319185256958,
|
|
"num_tokens": 2331661.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.1420262336730955,
|
|
"epoch": 1.0263516583371195,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000499815255189803,
|
|
"loss": 6.0517,
|
|
"mean_token_accuracy": 0.13892114236950875,
|
|
"num_tokens": 2341795.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.079546356201172,
|
|
"epoch": 1.0308950477055883,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998006584980922,
|
|
"loss": 5.9691,
|
|
"mean_token_accuracy": 0.13563258126378058,
|
|
"num_tokens": 2351322.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.145311260223389,
|
|
"epoch": 1.0354384370740572,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004997855071331034,
|
|
"loss": 6.0101,
|
|
"mean_token_accuracy": 0.1399790108203888,
|
|
"num_tokens": 2361111.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.131455659866333,
|
|
"epoch": 1.039981826442526,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004997698011322211,
|
|
"loss": 6.0829,
|
|
"mean_token_accuracy": 0.13193223252892494,
|
|
"num_tokens": 2371481.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.159214019775391,
|
|
"epoch": 1.044525215810995,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000499753540534198,
|
|
"loss": 5.9465,
|
|
"mean_token_accuracy": 0.13461560755968094,
|
|
"num_tokens": 2382109.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.150908994674682,
|
|
"epoch": 1.049068605179464,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499736725379156,
|
|
"loss": 6.1083,
|
|
"mean_token_accuracy": 0.13818338736891747,
|
|
"num_tokens": 2392451.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.15005316734314,
|
|
"epoch": 1.0536119945479328,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997193557085844,
|
|
"loss": 5.9726,
|
|
"mean_token_accuracy": 0.1419435605406761,
|
|
"num_tokens": 2402043.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.23726077079773,
|
|
"epoch": 1.0581553839164017,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004997014315653413,
|
|
"loss": 6.1305,
|
|
"mean_token_accuracy": 0.13321863487362862,
|
|
"num_tokens": 2411730.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.111984491348267,
|
|
"epoch": 1.0626987732848705,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004996829529936527,
|
|
"loss": 6.0535,
|
|
"mean_token_accuracy": 0.1366070255637169,
|
|
"num_tokens": 2421881.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.155670213699341,
|
|
"epoch": 1.0672421626533395,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004996639200391126,
|
|
"loss": 6.0484,
|
|
"mean_token_accuracy": 0.14234526082873344,
|
|
"num_tokens": 2431702.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.079865741729736,
|
|
"epoch": 1.0717855520218083,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004996443327486831,
|
|
"loss": 5.9995,
|
|
"mean_token_accuracy": 0.13819927722215652,
|
|
"num_tokens": 2442327.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.146263885498047,
|
|
"epoch": 1.0763289413902772,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996241911706937,
|
|
"loss": 6.0363,
|
|
"mean_token_accuracy": 0.13991880789399147,
|
|
"num_tokens": 2452155.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.133799982070923,
|
|
"epoch": 1.080872330758746,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004996034953548418,
|
|
"loss": 6.0365,
|
|
"mean_token_accuracy": 0.1321781098842621,
|
|
"num_tokens": 2462632.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.138056516647339,
|
|
"epoch": 1.085415720127215,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004995822453521924,
|
|
"loss": 5.9675,
|
|
"mean_token_accuracy": 0.14258275032043458,
|
|
"num_tokens": 2472326.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.116362905502319,
|
|
"epoch": 1.0899591094956838,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004995604412151776,
|
|
"loss": 6.0471,
|
|
"mean_token_accuracy": 0.13381395041942595,
|
|
"num_tokens": 2483394.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.082976675033569,
|
|
"epoch": 1.0945024988641527,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499538082997597,
|
|
"loss": 6.0237,
|
|
"mean_token_accuracy": 0.13501874655485152,
|
|
"num_tokens": 2493863.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.173003053665161,
|
|
"epoch": 1.0990458882326215,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004995151707546172,
|
|
"loss": 6.0612,
|
|
"mean_token_accuracy": 0.12664901614189147,
|
|
"num_tokens": 2504565.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.11851019859314,
|
|
"epoch": 1.1035892776010905,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004994917045427721,
|
|
"loss": 6.1012,
|
|
"mean_token_accuracy": 0.13649318665266036,
|
|
"num_tokens": 2515323.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.114333629608154,
|
|
"epoch": 1.1081326669695593,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004994676844199621,
|
|
"loss": 6.042,
|
|
"mean_token_accuracy": 0.13800725415349008,
|
|
"num_tokens": 2526158.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.121361970901489,
|
|
"epoch": 1.1126760563380282,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004994431104454543,
|
|
"loss": 6.0556,
|
|
"mean_token_accuracy": 0.1371620051562786,
|
|
"num_tokens": 2536814.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.0955418109893795,
|
|
"epoch": 1.117219445706497,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004994179826798828,
|
|
"loss": 6.0267,
|
|
"mean_token_accuracy": 0.1353412687778473,
|
|
"num_tokens": 2547534.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.083361291885376,
|
|
"epoch": 1.121762835074966,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004993923011852477,
|
|
"loss": 6.071,
|
|
"mean_token_accuracy": 0.13434085100889206,
|
|
"num_tokens": 2558007.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.174566650390625,
|
|
"epoch": 1.1263062244434348,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004993660660249158,
|
|
"loss": 6.0718,
|
|
"mean_token_accuracy": 0.1386457212269306,
|
|
"num_tokens": 2569523.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.05771107673645,
|
|
"epoch": 1.1308496138119037,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004993392772636194,
|
|
"loss": 5.9552,
|
|
"mean_token_accuracy": 0.13412538096308707,
|
|
"num_tokens": 2580133.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.030849838256836,
|
|
"epoch": 1.1353930031803725,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004993119349674573,
|
|
"loss": 5.856,
|
|
"mean_token_accuracy": 0.1392548970878124,
|
|
"num_tokens": 2589994.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.008020687103271,
|
|
"epoch": 1.1399363925488415,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004992840392038938,
|
|
"loss": 6.0189,
|
|
"mean_token_accuracy": 0.13825450092554092,
|
|
"num_tokens": 2600675.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.147768020629883,
|
|
"epoch": 1.1444797819173103,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000499255590041759,
|
|
"loss": 6.0472,
|
|
"mean_token_accuracy": 0.12776985839009286,
|
|
"num_tokens": 2611468.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.028980398178101,
|
|
"epoch": 1.1490231712857792,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004992265875512485,
|
|
"loss": 5.8744,
|
|
"mean_token_accuracy": 0.14487501680850984,
|
|
"num_tokens": 2621920.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 5.9842596530914305,
|
|
"epoch": 1.153566560654248,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004991970318039228,
|
|
"loss": 6.0354,
|
|
"mean_token_accuracy": 0.13384448885917663,
|
|
"num_tokens": 2632235.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.039266777038574,
|
|
"epoch": 1.158109950022717,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004991669228727078,
|
|
"loss": 5.9684,
|
|
"mean_token_accuracy": 0.14427078366279603,
|
|
"num_tokens": 2642124.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.068850898742676,
|
|
"epoch": 1.1626533393911858,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004991362608318947,
|
|
"loss": 5.9384,
|
|
"mean_token_accuracy": 0.14294721484184264,
|
|
"num_tokens": 2652706.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 5.983570957183838,
|
|
"epoch": 1.1671967287596547,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004991050457571384,
|
|
"loss": 5.9937,
|
|
"mean_token_accuracy": 0.1419117584824562,
|
|
"num_tokens": 2663432.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.102989578247071,
|
|
"epoch": 1.1717401181281235,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004990732777254594,
|
|
"loss": 6.0254,
|
|
"mean_token_accuracy": 0.13163735792040826,
|
|
"num_tokens": 2673049.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.130951070785523,
|
|
"epoch": 1.1762835074965925,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004990409568152422,
|
|
"loss": 6.056,
|
|
"mean_token_accuracy": 0.1352098412811756,
|
|
"num_tokens": 2683857.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 5.973699998855591,
|
|
"epoch": 1.1808268968650613,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004990080831062352,
|
|
"loss": 5.9929,
|
|
"mean_token_accuracy": 0.1435897931456566,
|
|
"num_tokens": 2694304.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.127169609069824,
|
|
"epoch": 1.1853702862335302,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004989746566795512,
|
|
"loss": 6.0485,
|
|
"mean_token_accuracy": 0.13189298063516616,
|
|
"num_tokens": 2704381.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.148133134841919,
|
|
"epoch": 1.189913675601999,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004989406776176665,
|
|
"loss": 6.0702,
|
|
"mean_token_accuracy": 0.1281648263335228,
|
|
"num_tokens": 2716006.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.052830934524536,
|
|
"epoch": 1.194457064970468,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000498906146004421,
|
|
"loss": 6.0813,
|
|
"mean_token_accuracy": 0.13499361276626587,
|
|
"num_tokens": 2726012.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.102909851074219,
|
|
"epoch": 1.1990004543389368,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004988710619250181,
|
|
"loss": 6.0031,
|
|
"mean_token_accuracy": 0.13929674178361892,
|
|
"num_tokens": 2735532.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.080796909332276,
|
|
"epoch": 1.2035438437074057,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004988354254660244,
|
|
"loss": 6.0198,
|
|
"mean_token_accuracy": 0.14065880179405213,
|
|
"num_tokens": 2745172.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 5.985582876205444,
|
|
"epoch": 1.2080872330758745,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004987992367153691,
|
|
"loss": 5.9146,
|
|
"mean_token_accuracy": 0.14265609309077262,
|
|
"num_tokens": 2756211.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.039712905883789,
|
|
"epoch": 1.2126306224443435,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004987624957623443,
|
|
"loss": 6.1229,
|
|
"mean_token_accuracy": 0.133643551915884,
|
|
"num_tokens": 2767933.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.086614799499512,
|
|
"epoch": 1.2171740118128123,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004987252026976049,
|
|
"loss": 5.9834,
|
|
"mean_token_accuracy": 0.1415078692138195,
|
|
"num_tokens": 2778176.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.037263250350952,
|
|
"epoch": 1.2217174011812812,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004986873576131676,
|
|
"loss": 5.9931,
|
|
"mean_token_accuracy": 0.13973900973796843,
|
|
"num_tokens": 2788486.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.134681844711304,
|
|
"epoch": 1.22626079054975,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004986489606024116,
|
|
"loss": 6.0902,
|
|
"mean_token_accuracy": 0.13812331557273866,
|
|
"num_tokens": 2799549.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.029563140869141,
|
|
"epoch": 1.230804179918219,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004986100117600776,
|
|
"loss": 6.0134,
|
|
"mean_token_accuracy": 0.1400228500366211,
|
|
"num_tokens": 2810674.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.097317552566528,
|
|
"epoch": 1.235347569286688,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000498570511182268,
|
|
"loss": 5.9747,
|
|
"mean_token_accuracy": 0.13779623731970786,
|
|
"num_tokens": 2821011.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.012727117538452,
|
|
"epoch": 1.2398909586551567,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004985304589664465,
|
|
"loss": 5.9328,
|
|
"mean_token_accuracy": 0.14200753346085548,
|
|
"num_tokens": 2831245.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.065994548797607,
|
|
"epoch": 1.2444343480236255,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004984898552114381,
|
|
"loss": 5.9852,
|
|
"mean_token_accuracy": 0.1369895800948143,
|
|
"num_tokens": 2841397.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.114771318435669,
|
|
"epoch": 1.2489777373920945,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004984487000174283,
|
|
"loss": 6.0306,
|
|
"mean_token_accuracy": 0.13484285473823548,
|
|
"num_tokens": 2851971.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 5.974376487731933,
|
|
"epoch": 1.2535211267605635,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004984069934859637,
|
|
"loss": 5.9745,
|
|
"mean_token_accuracy": 0.14509348720312118,
|
|
"num_tokens": 2861882.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.04203724861145,
|
|
"epoch": 1.2580645161290323,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004983647357199509,
|
|
"loss": 6.0139,
|
|
"mean_token_accuracy": 0.14242928475141525,
|
|
"num_tokens": 2872888.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 5.974927806854248,
|
|
"epoch": 1.262607905497501,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004983219268236567,
|
|
"loss": 5.9679,
|
|
"mean_token_accuracy": 0.14272170662879943,
|
|
"num_tokens": 2883194.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.067883062362671,
|
|
"epoch": 1.26715129486597,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004982785669027079,
|
|
"loss": 5.9561,
|
|
"mean_token_accuracy": 0.14673515483736993,
|
|
"num_tokens": 2892672.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.952126836776733,
|
|
"epoch": 1.271694684234439,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004982346560640907,
|
|
"loss": 5.9802,
|
|
"mean_token_accuracy": 0.13936009705066682,
|
|
"num_tokens": 2902155.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.082441091537476,
|
|
"epoch": 1.2762380736029078,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004981901944161507,
|
|
"loss": 5.9298,
|
|
"mean_token_accuracy": 0.14501121267676353,
|
|
"num_tokens": 2911805.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 5.976729249954223,
|
|
"epoch": 1.2807814629713765,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004981451820685928,
|
|
"loss": 5.9421,
|
|
"mean_token_accuracy": 0.13730958774685859,
|
|
"num_tokens": 2922975.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.961244773864746,
|
|
"epoch": 1.2853248523398455,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004980996191324802,
|
|
"loss": 5.9758,
|
|
"mean_token_accuracy": 0.1435778558254242,
|
|
"num_tokens": 2932416.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.068545198440551,
|
|
"epoch": 1.2898682417083145,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004980535057202351,
|
|
"loss": 5.9246,
|
|
"mean_token_accuracy": 0.1430967390537262,
|
|
"num_tokens": 2943336.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.946337366104126,
|
|
"epoch": 1.2944116310767833,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004980068419456378,
|
|
"loss": 5.9097,
|
|
"mean_token_accuracy": 0.1428004875779152,
|
|
"num_tokens": 2953397.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.007153081893921,
|
|
"epoch": 1.2989550204452522,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004979596279238265,
|
|
"loss": 6.0119,
|
|
"mean_token_accuracy": 0.14089832603931426,
|
|
"num_tokens": 2963811.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.014495849609375,
|
|
"epoch": 1.303498409813721,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004979118637712969,
|
|
"loss": 5.9594,
|
|
"mean_token_accuracy": 0.13529563397169114,
|
|
"num_tokens": 2973309.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.009972476959229,
|
|
"epoch": 1.30804179918219,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004978635496059025,
|
|
"loss": 5.9054,
|
|
"mean_token_accuracy": 0.13381288573145866,
|
|
"num_tokens": 2984873.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.961966848373413,
|
|
"epoch": 1.3125851885506588,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004978146855468537,
|
|
"loss": 5.9804,
|
|
"mean_token_accuracy": 0.14162408858537673,
|
|
"num_tokens": 2995117.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 5.983718729019165,
|
|
"epoch": 1.3171285779191277,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004977652717147177,
|
|
"loss": 6.0084,
|
|
"mean_token_accuracy": 0.14052879959344863,
|
|
"num_tokens": 3005035.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.928684234619141,
|
|
"epoch": 1.3216719672875965,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004977153082314182,
|
|
"loss": 5.9406,
|
|
"mean_token_accuracy": 0.142349211871624,
|
|
"num_tokens": 3015091.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.030572557449341,
|
|
"epoch": 1.3262153566560655,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000497664795220235,
|
|
"loss": 6.023,
|
|
"mean_token_accuracy": 0.1426836997270584,
|
|
"num_tokens": 3025261.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.0664942264556885,
|
|
"epoch": 1.3307587460245343,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004976137328058042,
|
|
"loss": 6.0056,
|
|
"mean_token_accuracy": 0.1386928841471672,
|
|
"num_tokens": 3036198.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.9850257396697994,
|
|
"epoch": 1.3353021353930032,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000497562121114117,
|
|
"loss": 5.9862,
|
|
"mean_token_accuracy": 0.13778137564659118,
|
|
"num_tokens": 3047295.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.985382986068726,
|
|
"epoch": 1.339845524761472,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004975099602725202,
|
|
"loss": 5.941,
|
|
"mean_token_accuracy": 0.14224427342414855,
|
|
"num_tokens": 3057427.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.026016187667847,
|
|
"epoch": 1.344388914129941,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004974572504097154,
|
|
"loss": 5.9648,
|
|
"mean_token_accuracy": 0.14337582439184188,
|
|
"num_tokens": 3067774.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.028681468963623,
|
|
"epoch": 1.3489323034984098,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004974039916557591,
|
|
"loss": 5.9132,
|
|
"mean_token_accuracy": 0.14107456281781197,
|
|
"num_tokens": 3077286.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.127128171920776,
|
|
"epoch": 1.3534756928668787,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000497350184142062,
|
|
"loss": 5.9865,
|
|
"mean_token_accuracy": 0.13820576295256615,
|
|
"num_tokens": 3088195.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.905239915847778,
|
|
"epoch": 1.3580190822353475,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004972958280013886,
|
|
"loss": 5.8789,
|
|
"mean_token_accuracy": 0.13997682929039001,
|
|
"num_tokens": 3098158.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.993680334091186,
|
|
"epoch": 1.3625624716038165,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004972409233678575,
|
|
"loss": 5.9806,
|
|
"mean_token_accuracy": 0.14393218830227852,
|
|
"num_tokens": 3108456.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.955839443206787,
|
|
"epoch": 1.3671058609722853,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004971854703769401,
|
|
"loss": 5.8702,
|
|
"mean_token_accuracy": 0.14266545102000236,
|
|
"num_tokens": 3118366.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.0084840774536135,
|
|
"epoch": 1.3716492503407542,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004971294691654616,
|
|
"loss": 5.9273,
|
|
"mean_token_accuracy": 0.1385462649166584,
|
|
"num_tokens": 3128975.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.95463604927063,
|
|
"epoch": 1.376192639709223,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000497072919871599,
|
|
"loss": 5.9543,
|
|
"mean_token_accuracy": 0.1458298459649086,
|
|
"num_tokens": 3139299.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.927489757537842,
|
|
"epoch": 1.380736029077692,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004970158226348824,
|
|
"loss": 5.9245,
|
|
"mean_token_accuracy": 0.1443161353468895,
|
|
"num_tokens": 3149987.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.98699197769165,
|
|
"epoch": 1.3852794184461608,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004969581775961932,
|
|
"loss": 5.9591,
|
|
"mean_token_accuracy": 0.1391432523727417,
|
|
"num_tokens": 3160922.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.022416639328003,
|
|
"epoch": 1.3898228078146297,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004968999848977651,
|
|
"loss": 5.9974,
|
|
"mean_token_accuracy": 0.13953477665781974,
|
|
"num_tokens": 3172770.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.955699491500854,
|
|
"epoch": 1.3943661971830985,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004968412446831827,
|
|
"loss": 5.9561,
|
|
"mean_token_accuracy": 0.1413671687245369,
|
|
"num_tokens": 3183218.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.9225616455078125,
|
|
"epoch": 1.3989095865515675,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004967819570973816,
|
|
"loss": 5.7925,
|
|
"mean_token_accuracy": 0.1463052049279213,
|
|
"num_tokens": 3192737.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.961886882781982,
|
|
"epoch": 1.4034529759200365,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000496722122286648,
|
|
"loss": 6.0038,
|
|
"mean_token_accuracy": 0.13916029781103134,
|
|
"num_tokens": 3203802.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.000508880615234,
|
|
"epoch": 1.4079963652885052,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004966617403986185,
|
|
"loss": 5.9516,
|
|
"mean_token_accuracy": 0.14601216688752175,
|
|
"num_tokens": 3214232.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.122423076629639,
|
|
"epoch": 1.412539754656974,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004966008115822793,
|
|
"loss": 5.9903,
|
|
"mean_token_accuracy": 0.13714029863476754,
|
|
"num_tokens": 3225893.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.751972341537476,
|
|
"epoch": 1.417083144025443,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004965393359879662,
|
|
"loss": 5.8346,
|
|
"mean_token_accuracy": 0.14151518866419793,
|
|
"num_tokens": 3235118.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.064570665359497,
|
|
"epoch": 1.421626533393912,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004964773137673641,
|
|
"loss": 5.9124,
|
|
"mean_token_accuracy": 0.14920328110456466,
|
|
"num_tokens": 3245770.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.846122407913208,
|
|
"epoch": 1.4261699227623807,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004964147450735067,
|
|
"loss": 5.8949,
|
|
"mean_token_accuracy": 0.1421331949532032,
|
|
"num_tokens": 3255262.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.011489725112915,
|
|
"epoch": 1.4307133121308495,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000496351630060776,
|
|
"loss": 5.9839,
|
|
"mean_token_accuracy": 0.14037677198648452,
|
|
"num_tokens": 3265682.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.986287546157837,
|
|
"epoch": 1.4352567014993185,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004962879688849021,
|
|
"loss": 6.0301,
|
|
"mean_token_accuracy": 0.14520008191466333,
|
|
"num_tokens": 3275734.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.040392637252808,
|
|
"epoch": 1.4398000908677875,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004962237617029625,
|
|
"loss": 5.911,
|
|
"mean_token_accuracy": 0.13663775473833084,
|
|
"num_tokens": 3286186.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.068287038803101,
|
|
"epoch": 1.4443434802362562,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004961590086733821,
|
|
"loss": 6.0751,
|
|
"mean_token_accuracy": 0.13447670564055442,
|
|
"num_tokens": 3297543.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.971070957183838,
|
|
"epoch": 1.448886869604725,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004960937099559326,
|
|
"loss": 5.9092,
|
|
"mean_token_accuracy": 0.14828140735626222,
|
|
"num_tokens": 3307840.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.951090621948242,
|
|
"epoch": 1.453430258973194,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004960278657117321,
|
|
"loss": 5.9347,
|
|
"mean_token_accuracy": 0.14122510626912116,
|
|
"num_tokens": 3318502.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.972015810012818,
|
|
"epoch": 1.457973648341663,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004959614761032446,
|
|
"loss": 5.9201,
|
|
"mean_token_accuracy": 0.13903507143259047,
|
|
"num_tokens": 3329161.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.990231657028199,
|
|
"epoch": 1.4625170377101318,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00049589454129428,
|
|
"loss": 5.8633,
|
|
"mean_token_accuracy": 0.14469081088900565,
|
|
"num_tokens": 3338770.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.959870481491089,
|
|
"epoch": 1.4670604270786005,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004958270614499932,
|
|
"loss": 5.978,
|
|
"mean_token_accuracy": 0.1422998420894146,
|
|
"num_tokens": 3350807.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.944865751266479,
|
|
"epoch": 1.4716038164470695,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004957590367368841,
|
|
"loss": 5.8494,
|
|
"mean_token_accuracy": 0.14039375260472298,
|
|
"num_tokens": 3359943.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.028170776367188,
|
|
"epoch": 1.4761472058155385,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004956904673227968,
|
|
"loss": 5.9822,
|
|
"mean_token_accuracy": 0.14763114899396895,
|
|
"num_tokens": 3369560.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.860723352432251,
|
|
"epoch": 1.4806905951840073,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004956213533769195,
|
|
"loss": 5.9108,
|
|
"mean_token_accuracy": 0.14079108610749244,
|
|
"num_tokens": 3380541.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.023792171478272,
|
|
"epoch": 1.485233984552476,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004955516950697843,
|
|
"loss": 5.9863,
|
|
"mean_token_accuracy": 0.14471204578876495,
|
|
"num_tokens": 3392428.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.9620476245880125,
|
|
"epoch": 1.489777373920945,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004954814925732657,
|
|
"loss": 5.8748,
|
|
"mean_token_accuracy": 0.14238841086626053,
|
|
"num_tokens": 3403189.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.870425748825073,
|
|
"epoch": 1.494320763289414,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004954107460605816,
|
|
"loss": 5.842,
|
|
"mean_token_accuracy": 0.14517249315977096,
|
|
"num_tokens": 3413359.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.922900056838989,
|
|
"epoch": 1.4988641526578828,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004953394557062922,
|
|
"loss": 5.9348,
|
|
"mean_token_accuracy": 0.14482194930315018,
|
|
"num_tokens": 3424395.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.95229206085205,
|
|
"epoch": 1.5034075420263515,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004952676216862989,
|
|
"loss": 5.8959,
|
|
"mean_token_accuracy": 0.1466038256883621,
|
|
"num_tokens": 3434452.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.936011409759521,
|
|
"epoch": 1.5079509313948205,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004951952441778453,
|
|
"loss": 6.002,
|
|
"mean_token_accuracy": 0.14328072518110274,
|
|
"num_tokens": 3445075.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.050215101242065,
|
|
"epoch": 1.5124943207632895,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004951223233595158,
|
|
"loss": 5.8804,
|
|
"mean_token_accuracy": 0.15087313205003738,
|
|
"num_tokens": 3454862.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.9777295112609865,
|
|
"epoch": 1.5170377101317583,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000495048859411235,
|
|
"loss": 6.0501,
|
|
"mean_token_accuracy": 0.13969595432281495,
|
|
"num_tokens": 3466255.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.978166294097901,
|
|
"epoch": 1.521581099500227,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004949748525142681,
|
|
"loss": 5.8495,
|
|
"mean_token_accuracy": 0.14266801625490189,
|
|
"num_tokens": 3475810.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.911789226531982,
|
|
"epoch": 1.526124488868696,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004949003028512195,
|
|
"loss": 5.9071,
|
|
"mean_token_accuracy": 0.14459308981895447,
|
|
"num_tokens": 3486979.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.874098205566407,
|
|
"epoch": 1.530667878237165,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004948252106060334,
|
|
"loss": 5.8235,
|
|
"mean_token_accuracy": 0.1440264791250229,
|
|
"num_tokens": 3496678.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.995673894882202,
|
|
"epoch": 1.5352112676056338,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004947495759639923,
|
|
"loss": 5.9194,
|
|
"mean_token_accuracy": 0.14006965309381486,
|
|
"num_tokens": 3507330.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.7853700637817385,
|
|
"epoch": 1.5397546569741025,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004946733991117171,
|
|
"loss": 5.8492,
|
|
"mean_token_accuracy": 0.14746353328227996,
|
|
"num_tokens": 3516781.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.004105186462402,
|
|
"epoch": 1.5442980463425715,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004945966802371668,
|
|
"loss": 5.9066,
|
|
"mean_token_accuracy": 0.14546382948756217,
|
|
"num_tokens": 3526936.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.894688463211059,
|
|
"epoch": 1.5488414357110405,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004945194195296374,
|
|
"loss": 5.8427,
|
|
"mean_token_accuracy": 0.14690589904785156,
|
|
"num_tokens": 3537319.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.965080738067627,
|
|
"epoch": 1.5533848250795095,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000494441617179762,
|
|
"loss": 5.889,
|
|
"mean_token_accuracy": 0.14221190959215163,
|
|
"num_tokens": 3547448.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.873220205307007,
|
|
"epoch": 1.5579282144479782,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004943632733795104,
|
|
"loss": 5.8481,
|
|
"mean_token_accuracy": 0.14411542043089867,
|
|
"num_tokens": 3558939.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.913844060897827,
|
|
"epoch": 1.562471603816447,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004942843883221879,
|
|
"loss": 5.9341,
|
|
"mean_token_accuracy": 0.1473378099501133,
|
|
"num_tokens": 3569501.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.912490558624268,
|
|
"epoch": 1.567014993184916,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004942049622024357,
|
|
"loss": 5.9162,
|
|
"mean_token_accuracy": 0.14080541655421258,
|
|
"num_tokens": 3579262.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.953363609313965,
|
|
"epoch": 1.571558382553385,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004941249952162298,
|
|
"loss": 5.9164,
|
|
"mean_token_accuracy": 0.14955995380878448,
|
|
"num_tokens": 3589894.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.906908178329468,
|
|
"epoch": 1.5761017719218537,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004940444875608807,
|
|
"loss": 5.8796,
|
|
"mean_token_accuracy": 0.14327764958143235,
|
|
"num_tokens": 3600011.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.920686388015747,
|
|
"epoch": 1.5806451612903225,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004939634394350332,
|
|
"loss": 5.8318,
|
|
"mean_token_accuracy": 0.14824839904904366,
|
|
"num_tokens": 3609832.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.868525600433349,
|
|
"epoch": 1.5851885506587915,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004938818510386654,
|
|
"loss": 5.7622,
|
|
"mean_token_accuracy": 0.15004636198282242,
|
|
"num_tokens": 3620196.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.908840179443359,
|
|
"epoch": 1.5897319400272605,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004937997225730886,
|
|
"loss": 5.8677,
|
|
"mean_token_accuracy": 0.1446698360145092,
|
|
"num_tokens": 3630488.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.907616424560547,
|
|
"epoch": 1.5942753293957292,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004937170542409464,
|
|
"loss": 5.8445,
|
|
"mean_token_accuracy": 0.15007875561714173,
|
|
"num_tokens": 3640449.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.922990465164185,
|
|
"epoch": 1.598818718764198,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000493633846246215,
|
|
"loss": 5.9154,
|
|
"mean_token_accuracy": 0.14311289340257644,
|
|
"num_tokens": 3650728.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.943247747421265,
|
|
"epoch": 1.603362108132667,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004935500987942018,
|
|
"loss": 5.8803,
|
|
"mean_token_accuracy": 0.14469165802001954,
|
|
"num_tokens": 3659739.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.981834411621094,
|
|
"epoch": 1.607905497501136,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004934658120915452,
|
|
"loss": 5.9898,
|
|
"mean_token_accuracy": 0.13959717378020287,
|
|
"num_tokens": 3671071.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.886190128326416,
|
|
"epoch": 1.6124488868696047,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004933809863462142,
|
|
"loss": 5.899,
|
|
"mean_token_accuracy": 0.14077742993831635,
|
|
"num_tokens": 3681289.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.8254108905792235,
|
|
"epoch": 1.6169922762380735,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004932956217675082,
|
|
"loss": 5.8237,
|
|
"mean_token_accuracy": 0.1419857293367386,
|
|
"num_tokens": 3691344.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.910706806182861,
|
|
"epoch": 1.6215356656065425,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004932097185660555,
|
|
"loss": 5.8268,
|
|
"mean_token_accuracy": 0.14990733712911605,
|
|
"num_tokens": 3702040.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.804446792602539,
|
|
"epoch": 1.6260790549750115,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004931232769538138,
|
|
"loss": 5.8653,
|
|
"mean_token_accuracy": 0.1472170442342758,
|
|
"num_tokens": 3711677.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.00853590965271,
|
|
"epoch": 1.6306224443434802,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004930362971440692,
|
|
"loss": 5.8915,
|
|
"mean_token_accuracy": 0.14570934921503068,
|
|
"num_tokens": 3721911.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.887213802337646,
|
|
"epoch": 1.635165833711949,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004929487793514358,
|
|
"loss": 5.8404,
|
|
"mean_token_accuracy": 0.15700444281101228,
|
|
"num_tokens": 3731817.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.898286199569702,
|
|
"epoch": 1.639709223080418,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000492860723791855,
|
|
"loss": 5.8559,
|
|
"mean_token_accuracy": 0.14812351018190384,
|
|
"num_tokens": 3742052.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.916392612457275,
|
|
"epoch": 1.644252612448887,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004927721306825952,
|
|
"loss": 5.8246,
|
|
"mean_token_accuracy": 0.14850612059235574,
|
|
"num_tokens": 3751630.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.85957612991333,
|
|
"epoch": 1.6487960018173557,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004926830002422509,
|
|
"loss": 5.7883,
|
|
"mean_token_accuracy": 0.14857261329889299,
|
|
"num_tokens": 3761858.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.8779035091400145,
|
|
"epoch": 1.6533393911858245,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000492593332690743,
|
|
"loss": 5.8808,
|
|
"mean_token_accuracy": 0.1462581604719162,
|
|
"num_tokens": 3771679.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.983050107955933,
|
|
"epoch": 1.6578827805542935,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004925031282493167,
|
|
"loss": 5.8485,
|
|
"mean_token_accuracy": 0.14935702979564666,
|
|
"num_tokens": 3782777.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.779914617538452,
|
|
"epoch": 1.6624261699227625,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004924123871405428,
|
|
"loss": 5.7653,
|
|
"mean_token_accuracy": 0.15233860909938812,
|
|
"num_tokens": 3792863.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.8145787715911865,
|
|
"epoch": 1.6669695592912313,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004923211095883161,
|
|
"loss": 5.8622,
|
|
"mean_token_accuracy": 0.14865206852555274,
|
|
"num_tokens": 3802807.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.873548316955566,
|
|
"epoch": 1.6715129486597,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004922292958178546,
|
|
"loss": 5.7348,
|
|
"mean_token_accuracy": 0.15383028537034987,
|
|
"num_tokens": 3812284.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.887331914901734,
|
|
"epoch": 1.676056338028169,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004921369460557,
|
|
"loss": 5.8734,
|
|
"mean_token_accuracy": 0.14546664506196977,
|
|
"num_tokens": 3823274.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.917618370056152,
|
|
"epoch": 1.680599727396638,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004920440605297159,
|
|
"loss": 5.8205,
|
|
"mean_token_accuracy": 0.1530941680073738,
|
|
"num_tokens": 3833768.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.82075252532959,
|
|
"epoch": 1.6851431167651068,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000491950639469088,
|
|
"loss": 5.8264,
|
|
"mean_token_accuracy": 0.1495286628603935,
|
|
"num_tokens": 3843937.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.841720771789551,
|
|
"epoch": 1.6896865061335755,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004918566831043238,
|
|
"loss": 5.7541,
|
|
"mean_token_accuracy": 0.15494302958250045,
|
|
"num_tokens": 3855072.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.852200174331665,
|
|
"epoch": 1.6942298955020445,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000491762191667251,
|
|
"loss": 5.899,
|
|
"mean_token_accuracy": 0.1420705497264862,
|
|
"num_tokens": 3866160.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.91382098197937,
|
|
"epoch": 1.6987732848705135,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000491667165391018,
|
|
"loss": 5.8198,
|
|
"mean_token_accuracy": 0.1473158195614815,
|
|
"num_tokens": 3876863.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.841605949401855,
|
|
"epoch": 1.7033166742389823,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004915716045100927,
|
|
"loss": 5.8921,
|
|
"mean_token_accuracy": 0.1498577430844307,
|
|
"num_tokens": 3887754.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.968225002288818,
|
|
"epoch": 1.707860063607451,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004914755092602621,
|
|
"loss": 5.8684,
|
|
"mean_token_accuracy": 0.15568319112062454,
|
|
"num_tokens": 3897591.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.79333004951477,
|
|
"epoch": 1.71240345297592,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004913788798786314,
|
|
"loss": 5.9011,
|
|
"mean_token_accuracy": 0.14456812888383866,
|
|
"num_tokens": 3908250.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.038985109329223,
|
|
"epoch": 1.716946842344389,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004912817166036243,
|
|
"loss": 5.9343,
|
|
"mean_token_accuracy": 0.14368811398744583,
|
|
"num_tokens": 3919385.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.936837005615234,
|
|
"epoch": 1.7214902317128578,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004911840196749813,
|
|
"loss": 6.0785,
|
|
"mean_token_accuracy": 0.14032945781946182,
|
|
"num_tokens": 3929999.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.973236036300659,
|
|
"epoch": 1.7260336210813265,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004910857893337602,
|
|
"loss": 5.8643,
|
|
"mean_token_accuracy": 0.15427177473902703,
|
|
"num_tokens": 3941311.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.810912132263184,
|
|
"epoch": 1.7305770104497955,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004909870258223342,
|
|
"loss": 5.8158,
|
|
"mean_token_accuracy": 0.14748491644859313,
|
|
"num_tokens": 3951409.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.870548439025879,
|
|
"epoch": 1.7351203998182645,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004908877293843926,
|
|
"loss": 5.867,
|
|
"mean_token_accuracy": 0.1453956812620163,
|
|
"num_tokens": 3961089.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.914389753341675,
|
|
"epoch": 1.7396637891867333,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004907879002649395,
|
|
"loss": 5.764,
|
|
"mean_token_accuracy": 0.14613812118768693,
|
|
"num_tokens": 3970548.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.785353899002075,
|
|
"epoch": 1.744207178555202,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004906875387102935,
|
|
"loss": 5.7569,
|
|
"mean_token_accuracy": 0.14780572652816773,
|
|
"num_tokens": 3981826.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.979347085952758,
|
|
"epoch": 1.748750567923671,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004905866449680864,
|
|
"loss": 5.9131,
|
|
"mean_token_accuracy": 0.1467357024550438,
|
|
"num_tokens": 3992063.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.8702459812164305,
|
|
"epoch": 1.75329395729214,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004904852192872638,
|
|
"loss": 5.8519,
|
|
"mean_token_accuracy": 0.14846348389983177,
|
|
"num_tokens": 4002007.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.827515935897827,
|
|
"epoch": 1.757837346660609,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004903832619180834,
|
|
"loss": 5.7665,
|
|
"mean_token_accuracy": 0.15415919572114944,
|
|
"num_tokens": 4012087.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.825321817398072,
|
|
"epoch": 1.7623807360290777,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004902807731121147,
|
|
"loss": 5.8063,
|
|
"mean_token_accuracy": 0.1539552964270115,
|
|
"num_tokens": 4022862.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.952997779846191,
|
|
"epoch": 1.7669241253975465,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004901777531222389,
|
|
"loss": 5.972,
|
|
"mean_token_accuracy": 0.14374119713902472,
|
|
"num_tokens": 4034992.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.880403089523315,
|
|
"epoch": 1.7714675147660155,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004900742022026474,
|
|
"loss": 5.8525,
|
|
"mean_token_accuracy": 0.1453924223780632,
|
|
"num_tokens": 4045749.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.856936359405518,
|
|
"epoch": 1.7760109041344845,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004899701206088419,
|
|
"loss": 5.7837,
|
|
"mean_token_accuracy": 0.14623216465115546,
|
|
"num_tokens": 4055733.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.943859243392945,
|
|
"epoch": 1.7805542935029532,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004898655085976333,
|
|
"loss": 5.8698,
|
|
"mean_token_accuracy": 0.14584364071488382,
|
|
"num_tokens": 4066153.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.876303386688233,
|
|
"epoch": 1.785097682871422,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004897603664271413,
|
|
"loss": 5.8122,
|
|
"mean_token_accuracy": 0.15110410153865814,
|
|
"num_tokens": 4075737.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.836494112014771,
|
|
"epoch": 1.789641072239891,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000489654694356794,
|
|
"loss": 5.8644,
|
|
"mean_token_accuracy": 0.15041233897209166,
|
|
"num_tokens": 4085642.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.96187949180603,
|
|
"epoch": 1.79418446160836,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004895484926473265,
|
|
"loss": 5.8934,
|
|
"mean_token_accuracy": 0.15415116995573044,
|
|
"num_tokens": 4095930.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.879729413986206,
|
|
"epoch": 1.7987278509768287,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000489441761560781,
|
|
"loss": 5.7855,
|
|
"mean_token_accuracy": 0.15090516060590745,
|
|
"num_tokens": 4105707.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.780129861831665,
|
|
"epoch": 1.8032712403452975,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000489334501360506,
|
|
"loss": 5.806,
|
|
"mean_token_accuracy": 0.1504233419895172,
|
|
"num_tokens": 4115891.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.896237373352051,
|
|
"epoch": 1.8078146297137665,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004892267123111551,
|
|
"loss": 5.8615,
|
|
"mean_token_accuracy": 0.15289803296327592,
|
|
"num_tokens": 4125873.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.788020658493042,
|
|
"epoch": 1.8123580190822355,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004891183946786873,
|
|
"loss": 5.8067,
|
|
"mean_token_accuracy": 0.1504621237516403,
|
|
"num_tokens": 4137162.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.849090909957885,
|
|
"epoch": 1.8169014084507042,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004890095487303655,
|
|
"loss": 5.7702,
|
|
"mean_token_accuracy": 0.14995464831590652,
|
|
"num_tokens": 4147431.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.8169014084507042,
|
|
"eval_entropy": 5.798743722605151,
|
|
"eval_loss": 6.04840087890625,
|
|
"eval_mean_token_accuracy": 0.14510284125111825,
|
|
"eval_num_tokens": 4147431.0,
|
|
"eval_runtime": 2.7705,
|
|
"eval_samples_per_second": 1239.862,
|
|
"eval_steps_per_second": 155.208,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.8538463592529295,
|
|
"epoch": 1.821444797819173,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004889001747347562,
|
|
"loss": 5.7016,
|
|
"mean_token_accuracy": 0.15428239703178406,
|
|
"num_tokens": 4157993.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.782310867309571,
|
|
"epoch": 1.825988187187642,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000488790272961729,
|
|
"loss": 5.7934,
|
|
"mean_token_accuracy": 0.1466740481555462,
|
|
"num_tokens": 4168834.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.798112392425537,
|
|
"epoch": 1.830531576556111,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004886798436824556,
|
|
"loss": 5.7602,
|
|
"mean_token_accuracy": 0.15480059683322905,
|
|
"num_tokens": 4178169.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.9378913879394535,
|
|
"epoch": 1.8350749659245797,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004885688871694091,
|
|
"loss": 5.8643,
|
|
"mean_token_accuracy": 0.14928691685199738,
|
|
"num_tokens": 4187817.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.86795482635498,
|
|
"epoch": 1.8396183552930485,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004884574036963639,
|
|
"loss": 5.8506,
|
|
"mean_token_accuracy": 0.14527582824230195,
|
|
"num_tokens": 4197577.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.883553504943848,
|
|
"epoch": 1.8441617446615175,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004883453935383941,
|
|
"loss": 5.8191,
|
|
"mean_token_accuracy": 0.15160404443740844,
|
|
"num_tokens": 4208234.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.775835227966309,
|
|
"epoch": 1.8487051340299865,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004882328569718738,
|
|
"loss": 5.7836,
|
|
"mean_token_accuracy": 0.1517630860209465,
|
|
"num_tokens": 4218562.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.899999475479126,
|
|
"epoch": 1.8532485233984552,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004881197942744758,
|
|
"loss": 5.9088,
|
|
"mean_token_accuracy": 0.15091830641031265,
|
|
"num_tokens": 4229202.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.8124762058258055,
|
|
"epoch": 1.857791912766924,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004880062057251711,
|
|
"loss": 5.678,
|
|
"mean_token_accuracy": 0.15422003269195556,
|
|
"num_tokens": 4238765.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.8610889434814455,
|
|
"epoch": 1.862335302135393,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000487892091604228,
|
|
"loss": 5.81,
|
|
"mean_token_accuracy": 0.14024123027920724,
|
|
"num_tokens": 4249453.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.743572092056274,
|
|
"epoch": 1.866878691503862,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048777745219321186,
|
|
"loss": 5.7194,
|
|
"mean_token_accuracy": 0.14468740597367286,
|
|
"num_tokens": 4259738.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.831390285491944,
|
|
"epoch": 1.8714220808723308,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00048766228777498405,
|
|
"loss": 5.9026,
|
|
"mean_token_accuracy": 0.14582569003105164,
|
|
"num_tokens": 4269405.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.906390714645386,
|
|
"epoch": 1.8759654702407995,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048754659863370125,
|
|
"loss": 5.7288,
|
|
"mean_token_accuracy": 0.14663628339767457,
|
|
"num_tokens": 4279821.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.712376976013184,
|
|
"epoch": 1.8805088596092685,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048743038505481503,
|
|
"loss": 5.747,
|
|
"mean_token_accuracy": 0.14767933040857315,
|
|
"num_tokens": 4290625.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.834017515182495,
|
|
"epoch": 1.8850522489777375,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004873136473250708,
|
|
"loss": 5.8011,
|
|
"mean_token_accuracy": 0.14723728150129317,
|
|
"num_tokens": 4301327.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.735596513748169,
|
|
"epoch": 1.8895956383462063,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048719638573250715,
|
|
"loss": 5.6988,
|
|
"mean_token_accuracy": 0.1526697278022766,
|
|
"num_tokens": 4311427.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.865527486801147,
|
|
"epoch": 1.894139027714675,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004870786005664555,
|
|
"loss": 5.8037,
|
|
"mean_token_accuracy": 0.14820494502782822,
|
|
"num_tokens": 4322258.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.896405649185181,
|
|
"epoch": 1.898682417083144,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004869602921175392,
|
|
"loss": 5.9037,
|
|
"mean_token_accuracy": 0.14564540088176728,
|
|
"num_tokens": 4332583.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.855921506881714,
|
|
"epoch": 1.903225806451613,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004868414606776724,
|
|
"loss": 5.9221,
|
|
"mean_token_accuracy": 0.14587701708078385,
|
|
"num_tokens": 4343027.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.971296787261963,
|
|
"epoch": 1.9077691958200818,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048672210654005996,
|
|
"loss": 5.9664,
|
|
"mean_token_accuracy": 0.13993714973330498,
|
|
"num_tokens": 4353234.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.78577356338501,
|
|
"epoch": 1.9123125851885505,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048660222999919616,
|
|
"loss": 5.7552,
|
|
"mean_token_accuracy": 0.15642950534820557,
|
|
"num_tokens": 4363074.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.752347373962403,
|
|
"epoch": 1.9168559745570195,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004864818313508647,
|
|
"loss": 5.7488,
|
|
"mean_token_accuracy": 0.1524498701095581,
|
|
"num_tokens": 4373303.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.875413036346435,
|
|
"epoch": 1.9213993639254885,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004863609108921372,
|
|
"loss": 5.758,
|
|
"mean_token_accuracy": 0.15022070854902267,
|
|
"num_tokens": 4382310.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.830654621124268,
|
|
"epoch": 1.9259427532939573,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004862394689213728,
|
|
"loss": 5.8264,
|
|
"mean_token_accuracy": 0.1505594179034233,
|
|
"num_tokens": 4393151.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.844172525405884,
|
|
"epoch": 1.930486142662426,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004861175057382175,
|
|
"loss": 5.7469,
|
|
"mean_token_accuracy": 0.15488230288028718,
|
|
"num_tokens": 4403036.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.8233729839324955,
|
|
"epoch": 1.935029532030895,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004859950216436034,
|
|
"loss": 5.7512,
|
|
"mean_token_accuracy": 0.14788225442171096,
|
|
"num_tokens": 4413703.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.837303209304809,
|
|
"epoch": 1.939572921399364,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004858720169397478,
|
|
"loss": 5.8395,
|
|
"mean_token_accuracy": 0.14474900364875792,
|
|
"num_tokens": 4424208.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.8106992721557615,
|
|
"epoch": 1.9441163107678328,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004857484919301525,
|
|
"loss": 5.8347,
|
|
"mean_token_accuracy": 0.14674064815044402,
|
|
"num_tokens": 4434077.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.826570606231689,
|
|
"epoch": 1.9486597001363015,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004856244469196034,
|
|
"loss": 5.7411,
|
|
"mean_token_accuracy": 0.15387579798698425,
|
|
"num_tokens": 4444344.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.7635636806488035,
|
|
"epoch": 1.9532030895047705,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000485499882214169,
|
|
"loss": 5.6898,
|
|
"mean_token_accuracy": 0.157929727435112,
|
|
"num_tokens": 4453632.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.771280765533447,
|
|
"epoch": 1.9577464788732395,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004853747981212003,
|
|
"loss": 5.7936,
|
|
"mean_token_accuracy": 0.15230028629302977,
|
|
"num_tokens": 4464139.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.917464017868042,
|
|
"epoch": 1.9622898682417085,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004852491949493301,
|
|
"loss": 5.844,
|
|
"mean_token_accuracy": 0.15021804422140123,
|
|
"num_tokens": 4474006.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.7739355087280275,
|
|
"epoch": 1.9668332576101772,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004851230730084716,
|
|
"loss": 5.7715,
|
|
"mean_token_accuracy": 0.15631687343120576,
|
|
"num_tokens": 4484494.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.857206058502197,
|
|
"epoch": 1.971376646978646,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048499643260981816,
|
|
"loss": 5.8,
|
|
"mean_token_accuracy": 0.15579398348927498,
|
|
"num_tokens": 4494694.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.734886598587036,
|
|
"epoch": 1.975920036347115,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004848692740658424,
|
|
"loss": 5.8027,
|
|
"mean_token_accuracy": 0.15389519929885864,
|
|
"num_tokens": 4504575.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.839699220657349,
|
|
"epoch": 1.980463425715584,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004847415976902954,
|
|
"loss": 5.8051,
|
|
"mean_token_accuracy": 0.14770661741495134,
|
|
"num_tokens": 4515767.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.863554525375366,
|
|
"epoch": 1.9850068150840527,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048461340379820587,
|
|
"loss": 5.9073,
|
|
"mean_token_accuracy": 0.14469317942857743,
|
|
"num_tokens": 4526194.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.874572706222534,
|
|
"epoch": 1.9895502044525215,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048448469270587945,
|
|
"loss": 5.8377,
|
|
"mean_token_accuracy": 0.14753615111112595,
|
|
"num_tokens": 4536662.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.851900196075439,
|
|
"epoch": 1.9940935938209905,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00048435546473089814,
|
|
"loss": 5.7767,
|
|
"mean_token_accuracy": 0.15122774839401246,
|
|
"num_tokens": 4546749.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.776758193969727,
|
|
"epoch": 1.9986369831894595,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048422572019211896,
|
|
"loss": 5.703,
|
|
"mean_token_accuracy": 0.15339836329221726,
|
|
"num_tokens": 4556194.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.779335021972656,
|
|
"epoch": 2.0027260336210815,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048409545940967375,
|
|
"loss": 5.576,
|
|
"mean_token_accuracy": 0.15858013927936554,
|
|
"num_tokens": 4563927.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.776248121261597,
|
|
"epoch": 2.00726942298955,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000483964682704968,
|
|
"loss": 5.4486,
|
|
"mean_token_accuracy": 0.16479917168617247,
|
|
"num_tokens": 4574800.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.733392667770386,
|
|
"epoch": 2.011812812358019,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048383339040068024,
|
|
"loss": 5.5052,
|
|
"mean_token_accuracy": 0.1624051257967949,
|
|
"num_tokens": 4585810.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.806802272796631,
|
|
"epoch": 2.016356201726488,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004837015828207612,
|
|
"loss": 5.6278,
|
|
"mean_token_accuracy": 0.15674534142017366,
|
|
"num_tokens": 4596227.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.777090406417846,
|
|
"epoch": 2.020899591094957,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004835692602904329,
|
|
"loss": 5.4859,
|
|
"mean_token_accuracy": 0.16422860473394393,
|
|
"num_tokens": 4606435.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.76586217880249,
|
|
"epoch": 2.0254429804634255,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004834364231361882,
|
|
"loss": 5.5091,
|
|
"mean_token_accuracy": 0.15572037696838378,
|
|
"num_tokens": 4616130.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.777824401855469,
|
|
"epoch": 2.0299863698318945,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004833030716857894,
|
|
"loss": 5.5088,
|
|
"mean_token_accuracy": 0.16416359841823577,
|
|
"num_tokens": 4626529.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.775467109680176,
|
|
"epoch": 2.0345297592003635,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048316920626826807,
|
|
"loss": 5.5025,
|
|
"mean_token_accuracy": 0.16245093941688538,
|
|
"num_tokens": 4637327.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.759325981140137,
|
|
"epoch": 2.0390731485688325,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00048303482721392363,
|
|
"loss": 5.4708,
|
|
"mean_token_accuracy": 0.16145365685224533,
|
|
"num_tokens": 4647412.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.776975536346436,
|
|
"epoch": 2.043616537937301,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048289993485432325,
|
|
"loss": 5.5525,
|
|
"mean_token_accuracy": 0.15803057104349136,
|
|
"num_tokens": 4659095.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.723943281173706,
|
|
"epoch": 2.04815992730577,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048276452952230036,
|
|
"loss": 5.4795,
|
|
"mean_token_accuracy": 0.17061184346675873,
|
|
"num_tokens": 4669239.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.823205375671387,
|
|
"epoch": 2.052703316674239,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048262861155195407,
|
|
"loss": 5.5317,
|
|
"mean_token_accuracy": 0.1543350949883461,
|
|
"num_tokens": 4679199.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.710643100738525,
|
|
"epoch": 2.057246706042708,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004824921812786486,
|
|
"loss": 5.4819,
|
|
"mean_token_accuracy": 0.1558678813278675,
|
|
"num_tokens": 4689884.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.7122931480407715,
|
|
"epoch": 2.0617900954111765,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000482355239039012,
|
|
"loss": 5.5341,
|
|
"mean_token_accuracy": 0.15882748365402222,
|
|
"num_tokens": 4700271.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.693128299713135,
|
|
"epoch": 2.0663334847796455,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004822177851709356,
|
|
"loss": 5.4944,
|
|
"mean_token_accuracy": 0.16197725534439086,
|
|
"num_tokens": 4709880.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.725936937332153,
|
|
"epoch": 2.0708768741481145,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00048207982001357327,
|
|
"loss": 5.4663,
|
|
"mean_token_accuracy": 0.1579332634806633,
|
|
"num_tokens": 4720492.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.68801097869873,
|
|
"epoch": 2.0754202635165835,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004819413439073402,
|
|
"loss": 5.537,
|
|
"mean_token_accuracy": 0.1539834961295128,
|
|
"num_tokens": 4731213.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.793666934967041,
|
|
"epoch": 2.079963652885052,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004818023571939125,
|
|
"loss": 5.4712,
|
|
"mean_token_accuracy": 0.1569421648979187,
|
|
"num_tokens": 4741680.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.605185031890869,
|
|
"epoch": 2.084507042253521,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000481662860216226,
|
|
"loss": 5.3746,
|
|
"mean_token_accuracy": 0.17384990006685258,
|
|
"num_tokens": 4750805.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.6640784740448,
|
|
"epoch": 2.08905043162199,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048152285331847574,
|
|
"loss": 5.4417,
|
|
"mean_token_accuracy": 0.16408300250768662,
|
|
"num_tokens": 4760554.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.674292945861817,
|
|
"epoch": 2.093593820990459,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004813823368461148,
|
|
"loss": 5.3829,
|
|
"mean_token_accuracy": 0.16840588450431823,
|
|
"num_tokens": 4770528.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.661532211303711,
|
|
"epoch": 2.098137210358928,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004812413111458534,
|
|
"loss": 5.4174,
|
|
"mean_token_accuracy": 0.1606951981782913,
|
|
"num_tokens": 4781216.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.646615839004516,
|
|
"epoch": 2.1026805997273965,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004810997765656587,
|
|
"loss": 5.4912,
|
|
"mean_token_accuracy": 0.1671098828315735,
|
|
"num_tokens": 4790329.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.686977338790894,
|
|
"epoch": 2.1072239890958655,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00048095773345475294,
|
|
"loss": 5.4388,
|
|
"mean_token_accuracy": 0.1641346976161003,
|
|
"num_tokens": 4800393.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.707598876953125,
|
|
"epoch": 2.1117673784643345,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00048081518216361367,
|
|
"loss": 5.5388,
|
|
"mean_token_accuracy": 0.16286579817533492,
|
|
"num_tokens": 4810297.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.730789613723755,
|
|
"epoch": 2.1163107678328035,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004806721230439719,
|
|
"loss": 5.4454,
|
|
"mean_token_accuracy": 0.16205213963985443,
|
|
"num_tokens": 4819415.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.682396125793457,
|
|
"epoch": 2.120854157201272,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004805285564488117,
|
|
"loss": 5.5011,
|
|
"mean_token_accuracy": 0.1609583541750908,
|
|
"num_tokens": 4829958.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.691484546661377,
|
|
"epoch": 2.125397546569741,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004803844827323696,
|
|
"loss": 5.486,
|
|
"mean_token_accuracy": 0.16416887789964676,
|
|
"num_tokens": 4839591.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.799030828475952,
|
|
"epoch": 2.12994093593821,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004802399022501329,
|
|
"loss": 5.5559,
|
|
"mean_token_accuracy": 0.1574663907289505,
|
|
"num_tokens": 4850697.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.5817474842071535,
|
|
"epoch": 2.134484325306679,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004800948153588397,
|
|
"loss": 5.4699,
|
|
"mean_token_accuracy": 0.16589853763580323,
|
|
"num_tokens": 4862523.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.6902327060699465,
|
|
"epoch": 2.1390277146751475,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047994922241647746,
|
|
"loss": 5.397,
|
|
"mean_token_accuracy": 0.16300795674324037,
|
|
"num_tokens": 4873432.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.673041486740113,
|
|
"epoch": 2.1435711040436165,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047980312378228244,
|
|
"loss": 5.518,
|
|
"mean_token_accuracy": 0.15471220016479492,
|
|
"num_tokens": 4884199.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.730643081665039,
|
|
"epoch": 2.1481144934120855,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047965651981673823,
|
|
"loss": 5.4789,
|
|
"mean_token_accuracy": 0.16016445010900499,
|
|
"num_tokens": 4894734.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.643956184387207,
|
|
"epoch": 2.1526578827805545,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047950941088157576,
|
|
"loss": 5.4034,
|
|
"mean_token_accuracy": 0.1633806273341179,
|
|
"num_tokens": 4905536.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.665775442123413,
|
|
"epoch": 2.157201272149023,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004793617973397715,
|
|
"loss": 5.4018,
|
|
"mean_token_accuracy": 0.17174910604953766,
|
|
"num_tokens": 4915176.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.67309250831604,
|
|
"epoch": 2.161744661517492,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00047921367955554745,
|
|
"loss": 5.4644,
|
|
"mean_token_accuracy": 0.16488925367593765,
|
|
"num_tokens": 4925937.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.645473146438599,
|
|
"epoch": 2.166288050885961,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004790650578943693,
|
|
"loss": 5.4695,
|
|
"mean_token_accuracy": 0.16120921671390534,
|
|
"num_tokens": 4937272.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.632696533203125,
|
|
"epoch": 2.17083144025443,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00047891593272294636,
|
|
"loss": 5.5763,
|
|
"mean_token_accuracy": 0.1590942397713661,
|
|
"num_tokens": 4947798.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.696415376663208,
|
|
"epoch": 2.1753748296228985,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047876630440923,
|
|
"loss": 5.5712,
|
|
"mean_token_accuracy": 0.15393886268138884,
|
|
"num_tokens": 4958150.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.698954343795776,
|
|
"epoch": 2.1799182189913675,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004786161733224134,
|
|
"loss": 5.4487,
|
|
"mean_token_accuracy": 0.1594776451587677,
|
|
"num_tokens": 4968588.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.663316869735718,
|
|
"epoch": 2.1844616083598365,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047846553983293006,
|
|
"loss": 5.5337,
|
|
"mean_token_accuracy": 0.16152950376272202,
|
|
"num_tokens": 4979169.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.752840995788574,
|
|
"epoch": 2.1890049977283055,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047831440431245307,
|
|
"loss": 5.612,
|
|
"mean_token_accuracy": 0.15289975255727767,
|
|
"num_tokens": 4990543.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.596266984939575,
|
|
"epoch": 2.193548387096774,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00047816276713389424,
|
|
"loss": 5.3947,
|
|
"mean_token_accuracy": 0.15945702493190766,
|
|
"num_tokens": 5000546.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.663093233108521,
|
|
"epoch": 2.198091776465243,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004780106286714034,
|
|
"loss": 5.5163,
|
|
"mean_token_accuracy": 0.16216596364974975,
|
|
"num_tokens": 5011092.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.6491169929504395,
|
|
"epoch": 2.202635165833712,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000477857989300367,
|
|
"loss": 5.4947,
|
|
"mean_token_accuracy": 0.16076005548238753,
|
|
"num_tokens": 5021317.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.694098377227784,
|
|
"epoch": 2.207178555202181,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047770484939740757,
|
|
"loss": 5.3969,
|
|
"mean_token_accuracy": 0.16764315515756606,
|
|
"num_tokens": 5031120.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.648887586593628,
|
|
"epoch": 2.2117219445706495,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004775512093403824,
|
|
"loss": 5.5462,
|
|
"mean_token_accuracy": 0.1566228985786438,
|
|
"num_tokens": 5040768.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.678833961486816,
|
|
"epoch": 2.2162653339391185,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00047739706950838327,
|
|
"loss": 5.4971,
|
|
"mean_token_accuracy": 0.1628997728228569,
|
|
"num_tokens": 5050639.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.655918455123901,
|
|
"epoch": 2.2208087233075875,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047724243028173484,
|
|
"loss": 5.4374,
|
|
"mean_token_accuracy": 0.16694911271333696,
|
|
"num_tokens": 5060624.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.62668809890747,
|
|
"epoch": 2.2253521126760565,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047708729204199406,
|
|
"loss": 5.5167,
|
|
"mean_token_accuracy": 0.15899598151445388,
|
|
"num_tokens": 5071894.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.689900016784668,
|
|
"epoch": 2.229895502044525,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000476931655171949,
|
|
"loss": 5.4751,
|
|
"mean_token_accuracy": 0.16796932816505433,
|
|
"num_tokens": 5082453.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.69913387298584,
|
|
"epoch": 2.234438891412994,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004767755200556183,
|
|
"loss": 5.5617,
|
|
"mean_token_accuracy": 0.15929114818572998,
|
|
"num_tokens": 5093033.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.687612295150757,
|
|
"epoch": 2.238982280781463,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004766188870782497,
|
|
"loss": 5.5417,
|
|
"mean_token_accuracy": 0.16963887810707093,
|
|
"num_tokens": 5103908.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.69229302406311,
|
|
"epoch": 2.243525670149932,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00047646175662631967,
|
|
"loss": 5.5017,
|
|
"mean_token_accuracy": 0.16409752070903777,
|
|
"num_tokens": 5114302.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.608609914779663,
|
|
"epoch": 2.2480690595184005,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047630412908753183,
|
|
"loss": 5.4077,
|
|
"mean_token_accuracy": 0.17001562267541886,
|
|
"num_tokens": 5124055.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.675837469100952,
|
|
"epoch": 2.2526124488868695,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047614600485081657,
|
|
"loss": 5.5237,
|
|
"mean_token_accuracy": 0.1581522673368454,
|
|
"num_tokens": 5134496.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.691870307922363,
|
|
"epoch": 2.2571558382553385,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047598738430632963,
|
|
"loss": 5.5511,
|
|
"mean_token_accuracy": 0.16181253939867019,
|
|
"num_tokens": 5146008.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.647511911392212,
|
|
"epoch": 2.2616992276238075,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00047582826784545146,
|
|
"loss": 5.4431,
|
|
"mean_token_accuracy": 0.1717854768037796,
|
|
"num_tokens": 5156733.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.7254432201385494,
|
|
"epoch": 2.266242616992276,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047566865586078615,
|
|
"loss": 5.4999,
|
|
"mean_token_accuracy": 0.16349462866783143,
|
|
"num_tokens": 5166452.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.634735774993897,
|
|
"epoch": 2.270786006360745,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047550854874616047,
|
|
"loss": 5.5769,
|
|
"mean_token_accuracy": 0.16026249378919602,
|
|
"num_tokens": 5177984.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.796362590789795,
|
|
"epoch": 2.275329395729214,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047534794689662257,
|
|
"loss": 5.5637,
|
|
"mean_token_accuracy": 0.15804722011089326,
|
|
"num_tokens": 5187815.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.662720251083374,
|
|
"epoch": 2.279872785097683,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004751868507084418,
|
|
"loss": 5.522,
|
|
"mean_token_accuracy": 0.1579802542924881,
|
|
"num_tokens": 5198333.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.658054828643799,
|
|
"epoch": 2.284416174466152,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004750252605791068,
|
|
"loss": 5.4597,
|
|
"mean_token_accuracy": 0.17050717920064926,
|
|
"num_tokens": 5209068.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.637749052047729,
|
|
"epoch": 2.2889595638346205,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004748631769073254,
|
|
"loss": 5.4522,
|
|
"mean_token_accuracy": 0.16521517485380172,
|
|
"num_tokens": 5218986.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.607774448394776,
|
|
"epoch": 2.2935029532030895,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004747006000930227,
|
|
"loss": 5.4855,
|
|
"mean_token_accuracy": 0.1623449444770813,
|
|
"num_tokens": 5228242.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.673880338668823,
|
|
"epoch": 2.2980463425715585,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004745375305373409,
|
|
"loss": 5.4941,
|
|
"mean_token_accuracy": 0.16177093088626862,
|
|
"num_tokens": 5238772.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.633888149261475,
|
|
"epoch": 2.302589731940027,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047437396864263804,
|
|
"loss": 5.4693,
|
|
"mean_token_accuracy": 0.16467007696628572,
|
|
"num_tokens": 5248588.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.623581743240356,
|
|
"epoch": 2.307133121308496,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004742099148124866,
|
|
"loss": 5.4498,
|
|
"mean_token_accuracy": 0.16977469027042388,
|
|
"num_tokens": 5258593.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.738746976852417,
|
|
"epoch": 2.311676510676965,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004740453694516734,
|
|
"loss": 5.5555,
|
|
"mean_token_accuracy": 0.15645188391208648,
|
|
"num_tokens": 5269639.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.588364028930664,
|
|
"epoch": 2.316219900045434,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047388033296619746,
|
|
"loss": 5.5628,
|
|
"mean_token_accuracy": 0.16065715104341508,
|
|
"num_tokens": 5279428.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.7361993312835695,
|
|
"epoch": 2.320763289413903,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004737148057632701,
|
|
"loss": 5.4676,
|
|
"mean_token_accuracy": 0.1655708596110344,
|
|
"num_tokens": 5289986.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.563482046127319,
|
|
"epoch": 2.3253066787823715,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004735487882513131,
|
|
"loss": 5.422,
|
|
"mean_token_accuracy": 0.16048859804868698,
|
|
"num_tokens": 5300629.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.624101209640503,
|
|
"epoch": 2.3298500681508405,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004733822808399583,
|
|
"loss": 5.4688,
|
|
"mean_token_accuracy": 0.16383247971534728,
|
|
"num_tokens": 5310472.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.689055585861206,
|
|
"epoch": 2.3343934575193095,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047321528394004615,
|
|
"loss": 5.4999,
|
|
"mean_token_accuracy": 0.16311483830213547,
|
|
"num_tokens": 5320981.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.57588529586792,
|
|
"epoch": 2.3389368468877785,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00047304779796362485,
|
|
"loss": 5.4144,
|
|
"mean_token_accuracy": 0.16940876841545105,
|
|
"num_tokens": 5330949.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.59773621559143,
|
|
"epoch": 2.343480236256247,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047287982332394943,
|
|
"loss": 5.4522,
|
|
"mean_token_accuracy": 0.15855366736650467,
|
|
"num_tokens": 5340748.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.671061325073242,
|
|
"epoch": 2.348023625624716,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047271136043548057,
|
|
"loss": 5.6066,
|
|
"mean_token_accuracy": 0.1577791079878807,
|
|
"num_tokens": 5352923.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.621159791946411,
|
|
"epoch": 2.352567014993185,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004725424097138838,
|
|
"loss": 5.4287,
|
|
"mean_token_accuracy": 0.16791379302740098,
|
|
"num_tokens": 5361923.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.703397512435913,
|
|
"epoch": 2.357110404361654,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00047237297157602816,
|
|
"loss": 5.5385,
|
|
"mean_token_accuracy": 0.16827826499938964,
|
|
"num_tokens": 5372268.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.669145345687866,
|
|
"epoch": 2.3616537937301225,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004722030464399854,
|
|
"loss": 5.4358,
|
|
"mean_token_accuracy": 0.1655590459704399,
|
|
"num_tokens": 5382071.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.599398756027222,
|
|
"epoch": 2.3661971830985915,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000472032634725029,
|
|
"loss": 5.5129,
|
|
"mean_token_accuracy": 0.163224895298481,
|
|
"num_tokens": 5392139.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.694211149215699,
|
|
"epoch": 2.3707405724670605,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047186173685163273,
|
|
"loss": 5.5247,
|
|
"mean_token_accuracy": 0.16626497358083725,
|
|
"num_tokens": 5402868.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.617294597625732,
|
|
"epoch": 2.3752839618355295,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00047169035324147023,
|
|
"loss": 5.5259,
|
|
"mean_token_accuracy": 0.1652606412768364,
|
|
"num_tokens": 5413029.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.5882978439331055,
|
|
"epoch": 2.379827351203998,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00047151848431741363,
|
|
"loss": 5.4675,
|
|
"mean_token_accuracy": 0.164324614405632,
|
|
"num_tokens": 5423145.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.6815375804901125,
|
|
"epoch": 2.384370740572467,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004713461305035322,
|
|
"loss": 5.4933,
|
|
"mean_token_accuracy": 0.16378677934408187,
|
|
"num_tokens": 5433715.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.533969593048096,
|
|
"epoch": 2.388914129940936,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000471173292225092,
|
|
"loss": 5.3966,
|
|
"mean_token_accuracy": 0.16783206909894943,
|
|
"num_tokens": 5444565.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.664507865905762,
|
|
"epoch": 2.393457519309405,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047099996990855424,
|
|
"loss": 5.564,
|
|
"mean_token_accuracy": 0.16357118338346482,
|
|
"num_tokens": 5456955.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.67940583229065,
|
|
"epoch": 2.3980009086778735,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00047082616398157456,
|
|
"loss": 5.4963,
|
|
"mean_token_accuracy": 0.1606929361820221,
|
|
"num_tokens": 5466260.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.678316831588745,
|
|
"epoch": 2.4025442980463425,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004706518748730017,
|
|
"loss": 5.5533,
|
|
"mean_token_accuracy": 0.1584568589925766,
|
|
"num_tokens": 5478055.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.533257150650025,
|
|
"epoch": 2.4070876874148115,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004704771030128768,
|
|
"loss": 5.4014,
|
|
"mean_token_accuracy": 0.15996367260813713,
|
|
"num_tokens": 5487532.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.562580871582031,
|
|
"epoch": 2.4116310767832805,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00047030184883243203,
|
|
"loss": 5.5058,
|
|
"mean_token_accuracy": 0.1656596690416336,
|
|
"num_tokens": 5498197.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.656260013580322,
|
|
"epoch": 2.416174466151749,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00047012611276408964,
|
|
"loss": 5.3975,
|
|
"mean_token_accuracy": 0.16340485960245132,
|
|
"num_tokens": 5508320.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.488940572738647,
|
|
"epoch": 2.420717855520218,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004699498952414609,
|
|
"loss": 5.386,
|
|
"mean_token_accuracy": 0.17562676668167115,
|
|
"num_tokens": 5517392.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.5707038879394535,
|
|
"epoch": 2.425261244888687,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00046977319669934506,
|
|
"loss": 5.4478,
|
|
"mean_token_accuracy": 0.16371348202228547,
|
|
"num_tokens": 5527097.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.690631771087647,
|
|
"epoch": 2.429804634257156,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00046959601757372814,
|
|
"loss": 5.4562,
|
|
"mean_token_accuracy": 0.16645233631134032,
|
|
"num_tokens": 5536935.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.5774236679077145,
|
|
"epoch": 2.4343480236256245,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00046941835830178216,
|
|
"loss": 5.5162,
|
|
"mean_token_accuracy": 0.16189493536949157,
|
|
"num_tokens": 5547432.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.597412061691284,
|
|
"epoch": 2.4388914129940935,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00046924021932186357,
|
|
"loss": 5.434,
|
|
"mean_token_accuracy": 0.16296101212501526,
|
|
"num_tokens": 5558335.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.69479923248291,
|
|
"epoch": 2.4434348023625625,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046906160107351263,
|
|
"loss": 5.5146,
|
|
"mean_token_accuracy": 0.1591379687190056,
|
|
"num_tokens": 5568474.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.502760219573974,
|
|
"epoch": 2.4479781917310315,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00046888250399745217,
|
|
"loss": 5.5306,
|
|
"mean_token_accuracy": 0.1706564098596573,
|
|
"num_tokens": 5577310.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.678367328643799,
|
|
"epoch": 2.4525215810995,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00046870292853558646,
|
|
"loss": 5.4958,
|
|
"mean_token_accuracy": 0.16218772977590562,
|
|
"num_tokens": 5587791.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.741392707824707,
|
|
"epoch": 2.457064970467969,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046852287513099997,
|
|
"loss": 5.5677,
|
|
"mean_token_accuracy": 0.1556030698120594,
|
|
"num_tokens": 5598783.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.569098949432373,
|
|
"epoch": 2.461608359836438,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004683423442279566,
|
|
"loss": 5.4301,
|
|
"mean_token_accuracy": 0.1574172407388687,
|
|
"num_tokens": 5609143.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.633760261535644,
|
|
"epoch": 2.466151749204907,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004681613362718985,
|
|
"loss": 5.4613,
|
|
"mean_token_accuracy": 0.1616216480731964,
|
|
"num_tokens": 5619657.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.630750751495361,
|
|
"epoch": 2.470695138573376,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004679798517094447,
|
|
"loss": 5.4737,
|
|
"mean_token_accuracy": 0.16256414353847504,
|
|
"num_tokens": 5630233.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.534254503250122,
|
|
"epoch": 2.4752385279418445,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046779789098839043,
|
|
"loss": 5.4341,
|
|
"mean_token_accuracy": 0.16833728849887847,
|
|
"num_tokens": 5640878.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.665617895126343,
|
|
"epoch": 2.4797819173103135,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046761545455770555,
|
|
"loss": 5.6317,
|
|
"mean_token_accuracy": 0.15832689255475998,
|
|
"num_tokens": 5651335.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.643316745758057,
|
|
"epoch": 2.4843253066787825,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004674325428675339,
|
|
"loss": 5.4929,
|
|
"mean_token_accuracy": 0.16712235361337663,
|
|
"num_tokens": 5661981.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.594733762741089,
|
|
"epoch": 2.488868696047251,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004672491563691919,
|
|
"loss": 5.4664,
|
|
"mean_token_accuracy": 0.16800144612789153,
|
|
"num_tokens": 5671551.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.590738105773926,
|
|
"epoch": 2.49341208541572,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046706529551516753,
|
|
"loss": 5.4224,
|
|
"mean_token_accuracy": 0.16803559213876723,
|
|
"num_tokens": 5681185.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.583927965164184,
|
|
"epoch": 2.497955474784189,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046688096075911905,
|
|
"loss": 5.4799,
|
|
"mean_token_accuracy": 0.167832550406456,
|
|
"num_tokens": 5690967.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.682482576370239,
|
|
"epoch": 2.502498864152658,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004666961525558743,
|
|
"loss": 5.5578,
|
|
"mean_token_accuracy": 0.15779926180839537,
|
|
"num_tokens": 5702011.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.672723388671875,
|
|
"epoch": 2.507042253521127,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000466510871361429,
|
|
"loss": 5.4992,
|
|
"mean_token_accuracy": 0.16007170379161834,
|
|
"num_tokens": 5712062.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.571487760543823,
|
|
"epoch": 2.5115856428895955,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046632511763294615,
|
|
"loss": 5.4567,
|
|
"mean_token_accuracy": 0.1650468647480011,
|
|
"num_tokens": 5722828.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.653512048721313,
|
|
"epoch": 2.5161290322580645,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004661388918287546,
|
|
"loss": 5.5415,
|
|
"mean_token_accuracy": 0.16156239062547684,
|
|
"num_tokens": 5732552.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.692649173736572,
|
|
"epoch": 2.5206724216265335,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004659521944083481,
|
|
"loss": 5.4988,
|
|
"mean_token_accuracy": 0.16152568012475968,
|
|
"num_tokens": 5743283.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.58889479637146,
|
|
"epoch": 2.525215810995002,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004657650258323838,
|
|
"loss": 5.4804,
|
|
"mean_token_accuracy": 0.16362597346305846,
|
|
"num_tokens": 5753803.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.607439422607422,
|
|
"epoch": 2.529759200363471,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00046557738656268166,
|
|
"loss": 5.461,
|
|
"mean_token_accuracy": 0.17017837166786193,
|
|
"num_tokens": 5764050.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.618855047225952,
|
|
"epoch": 2.53430258973194,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000465389277062223,
|
|
"loss": 5.4213,
|
|
"mean_token_accuracy": 0.17354113161563872,
|
|
"num_tokens": 5774125.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.648334217071533,
|
|
"epoch": 2.538845979100409,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046520069779514917,
|
|
"loss": 5.5519,
|
|
"mean_token_accuracy": 0.16256483793258666,
|
|
"num_tokens": 5784762.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.690959930419922,
|
|
"epoch": 2.543389368468878,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004650116492267609,
|
|
"loss": 5.4731,
|
|
"mean_token_accuracy": 0.15543195307254792,
|
|
"num_tokens": 5795593.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.632666778564453,
|
|
"epoch": 2.5479327578373465,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004648221318235165,
|
|
"loss": 5.4807,
|
|
"mean_token_accuracy": 0.16186553835868836,
|
|
"num_tokens": 5805757.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.618985414505005,
|
|
"epoch": 2.5524761472058155,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004646321460530316,
|
|
"loss": 5.477,
|
|
"mean_token_accuracy": 0.16880376040935516,
|
|
"num_tokens": 5816197.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.515861463546753,
|
|
"epoch": 2.5570195365742845,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000464441692384077,
|
|
"loss": 5.484,
|
|
"mean_token_accuracy": 0.16270082890987397,
|
|
"num_tokens": 5826193.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.645616579055786,
|
|
"epoch": 2.561562925942753,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004642507712865784,
|
|
"loss": 5.4453,
|
|
"mean_token_accuracy": 0.16180969178676605,
|
|
"num_tokens": 5835934.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.575414037704467,
|
|
"epoch": 2.566106315311222,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046405938323161453,
|
|
"loss": 5.3715,
|
|
"mean_token_accuracy": 0.1692644700407982,
|
|
"num_tokens": 5846376.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.579668664932251,
|
|
"epoch": 2.570649704679691,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046386752869141624,
|
|
"loss": 5.5159,
|
|
"mean_token_accuracy": 0.1617092326283455,
|
|
"num_tokens": 5856741.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.617500591278076,
|
|
"epoch": 2.57519309404816,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004636752081393658,
|
|
"loss": 5.5385,
|
|
"mean_token_accuracy": 0.1621531665325165,
|
|
"num_tokens": 5867690.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.573635005950928,
|
|
"epoch": 2.579736483416629,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046348242204999487,
|
|
"loss": 5.4427,
|
|
"mean_token_accuracy": 0.16709815710783005,
|
|
"num_tokens": 5878522.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.593231630325318,
|
|
"epoch": 2.5842798727850975,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00046328917089898396,
|
|
"loss": 5.4585,
|
|
"mean_token_accuracy": 0.1684649333357811,
|
|
"num_tokens": 5888702.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.594196224212647,
|
|
"epoch": 2.5888232621535665,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004630954551631612,
|
|
"loss": 5.5187,
|
|
"mean_token_accuracy": 0.16131021380424498,
|
|
"num_tokens": 5900040.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.568029356002808,
|
|
"epoch": 2.5933666515220355,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046290127532050087,
|
|
"loss": 5.3928,
|
|
"mean_token_accuracy": 0.16215718686580657,
|
|
"num_tokens": 5911572.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.649325561523438,
|
|
"epoch": 2.5979100408905045,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004627066318501223,
|
|
"loss": 5.484,
|
|
"mean_token_accuracy": 0.16762312650680541,
|
|
"num_tokens": 5921106.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.540233564376831,
|
|
"epoch": 2.602453430258973,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046251152523228906,
|
|
"loss": 5.4415,
|
|
"mean_token_accuracy": 0.16675643771886825,
|
|
"num_tokens": 5930809.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.535998296737671,
|
|
"epoch": 2.606996819627442,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004623159559484074,
|
|
"loss": 5.5051,
|
|
"mean_token_accuracy": 0.1679874151945114,
|
|
"num_tokens": 5941828.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.6888713359832765,
|
|
"epoch": 2.611540208995911,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004621199244810248,
|
|
"loss": 5.5442,
|
|
"mean_token_accuracy": 0.1616669088602066,
|
|
"num_tokens": 5951774.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.622589206695556,
|
|
"epoch": 2.61608359836438,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046192343131382964,
|
|
"loss": 5.4613,
|
|
"mean_token_accuracy": 0.16773341298103334,
|
|
"num_tokens": 5961872.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.544913911819458,
|
|
"epoch": 2.620626987732849,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046172647693164926,
|
|
"loss": 5.387,
|
|
"mean_token_accuracy": 0.17031843960285187,
|
|
"num_tokens": 5972635.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.54814453125,
|
|
"epoch": 2.6251703771013175,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004615290618204489,
|
|
"loss": 5.4411,
|
|
"mean_token_accuracy": 0.16277870833873748,
|
|
"num_tokens": 5983042.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.637651252746582,
|
|
"epoch": 2.6297137664697865,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004613311864673309,
|
|
"loss": 5.6537,
|
|
"mean_token_accuracy": 0.14904230907559396,
|
|
"num_tokens": 5993859.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.735480260848999,
|
|
"epoch": 2.6342571558382555,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046113285136053286,
|
|
"loss": 5.5295,
|
|
"mean_token_accuracy": 0.16324471086263656,
|
|
"num_tokens": 6004210.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.65135760307312,
|
|
"epoch": 2.638800545206724,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000460934056989427,
|
|
"loss": 5.3804,
|
|
"mean_token_accuracy": 0.16609495133161545,
|
|
"num_tokens": 6014476.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.5331555843353275,
|
|
"epoch": 2.643343934575193,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004607348038445187,
|
|
"loss": 5.4401,
|
|
"mean_token_accuracy": 0.15993242114782333,
|
|
"num_tokens": 6024454.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.581075191497803,
|
|
"epoch": 2.647887323943662,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00046053509241744526,
|
|
"loss": 5.4713,
|
|
"mean_token_accuracy": 0.17017142474651337,
|
|
"num_tokens": 6035041.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.606084251403809,
|
|
"epoch": 2.652430713312131,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00046033492320097483,
|
|
"loss": 5.4586,
|
|
"mean_token_accuracy": 0.17169903963804245,
|
|
"num_tokens": 6045005.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.65831184387207,
|
|
"epoch": 2.6569741026806,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004601342966890049,
|
|
"loss": 5.426,
|
|
"mean_token_accuracy": 0.17063144892454146,
|
|
"num_tokens": 6054973.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.569364929199219,
|
|
"epoch": 2.6615174920490685,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045993321337656163,
|
|
"loss": 5.4,
|
|
"mean_token_accuracy": 0.16538814455270767,
|
|
"num_tokens": 6065857.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.584576749801636,
|
|
"epoch": 2.6660608814175375,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000459731673759798,
|
|
"loss": 5.4347,
|
|
"mean_token_accuracy": 0.168292573094368,
|
|
"num_tokens": 6075568.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.6245733261108395,
|
|
"epoch": 2.6706042707860065,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00045952967833599303,
|
|
"loss": 5.5432,
|
|
"mean_token_accuracy": 0.1643181711435318,
|
|
"num_tokens": 6086103.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.609689474105835,
|
|
"epoch": 2.675147660154475,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004593272276035502,
|
|
"loss": 5.4723,
|
|
"mean_token_accuracy": 0.16511984616518022,
|
|
"num_tokens": 6096983.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.5687517642974855,
|
|
"epoch": 2.679691049522944,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0004591243220619969,
|
|
"loss": 5.4509,
|
|
"mean_token_accuracy": 0.16505469083786012,
|
|
"num_tokens": 6106518.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.609748601913452,
|
|
"epoch": 2.684234438891413,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000458920962211982,
|
|
"loss": 5.5302,
|
|
"mean_token_accuracy": 0.16398297548294066,
|
|
"num_tokens": 6118034.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.6170238018035885,
|
|
"epoch": 2.688777828259882,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004587171485552759,
|
|
"loss": 5.4812,
|
|
"mean_token_accuracy": 0.16860826909542084,
|
|
"num_tokens": 6128177.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.583043813705444,
|
|
"epoch": 2.693321217628351,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00045851288159476854,
|
|
"loss": 5.5264,
|
|
"mean_token_accuracy": 0.15912237763404846,
|
|
"num_tokens": 6138401.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.627243518829346,
|
|
"epoch": 2.6978646069968195,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004583081618344684,
|
|
"loss": 5.4292,
|
|
"mean_token_accuracy": 0.16771573275327684,
|
|
"num_tokens": 6148475.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.59838833808899,
|
|
"epoch": 2.7024079963652885,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000458102989779501,
|
|
"loss": 5.5747,
|
|
"mean_token_accuracy": 0.16113339066505433,
|
|
"num_tokens": 6159457.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.731915092468261,
|
|
"epoch": 2.7069513857337575,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004578973659361081,
|
|
"loss": 5.5392,
|
|
"mean_token_accuracy": 0.15799466669559478,
|
|
"num_tokens": 6169576.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.649990749359131,
|
|
"epoch": 2.711494775102226,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00045769129081164596,
|
|
"loss": 5.4841,
|
|
"mean_token_accuracy": 0.16854571998119355,
|
|
"num_tokens": 6179292.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.518716049194336,
|
|
"epoch": 2.716038164470695,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004574847649145845,
|
|
"loss": 5.4587,
|
|
"mean_token_accuracy": 0.16538277864456177,
|
|
"num_tokens": 6190824.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.635785722732544,
|
|
"epoch": 2.720581553839164,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000457277788754506,
|
|
"loss": 5.4653,
|
|
"mean_token_accuracy": 0.16698568016290666,
|
|
"num_tokens": 6201830.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.603476905822754,
|
|
"epoch": 2.725124943207633,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00045707036284210344,
|
|
"loss": 5.518,
|
|
"mean_token_accuracy": 0.1614304631948471,
|
|
"num_tokens": 6213063.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.725124943207633,
|
|
"eval_entropy": 5.448676344405773,
|
|
"eval_loss": 5.910400867462158,
|
|
"eval_mean_token_accuracy": 0.15391066744923593,
|
|
"eval_num_tokens": 6213063.0,
|
|
"eval_runtime": 2.5754,
|
|
"eval_samples_per_second": 1333.767,
|
|
"eval_steps_per_second": 166.964,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 1000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 8992180539648000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|