2046 lines
54 KiB
JSON
2046 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.9086778736937755,
|
|
"eval_steps": 1000,
|
|
"global_step": 1000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 3014967103488000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|