4057 lines
109 KiB
JSON
4057 lines
109 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.162249864791779,
|
|
"eval_steps": 1000,
|
|
"global_step": 2000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.692032432556152,
|
|
"epoch": 0.005408328826392645,
|
|
"grad_norm": 13.4375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8497,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 10948.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691972923278808,
|
|
"epoch": 0.01081665765278529,
|
|
"grad_norm": 13.4375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7725,
|
|
"mean_token_accuracy": 0.00017152659129351377,
|
|
"num_tokens": 21455.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.691221618652344,
|
|
"epoch": 0.016224986479177934,
|
|
"grad_norm": 10.125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.508,
|
|
"mean_token_accuracy": 0.017251307470723988,
|
|
"num_tokens": 32377.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.68411626815796,
|
|
"epoch": 0.02163331530557058,
|
|
"grad_norm": 6.6875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.1253,
|
|
"mean_token_accuracy": 0.036476511135697366,
|
|
"num_tokens": 42276.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.645281887054443,
|
|
"epoch": 0.02704164413196322,
|
|
"grad_norm": 4.21875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.8249,
|
|
"mean_token_accuracy": 0.03992591369897127,
|
|
"num_tokens": 55049.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.584981632232665,
|
|
"epoch": 0.03244997295835587,
|
|
"grad_norm": 3.203125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.6624,
|
|
"mean_token_accuracy": 0.03540853708982468,
|
|
"num_tokens": 66792.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.57096061706543,
|
|
"epoch": 0.03785830178474851,
|
|
"grad_norm": 2.9375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.5971,
|
|
"mean_token_accuracy": 0.03990415818989277,
|
|
"num_tokens": 78208.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.563229370117188,
|
|
"epoch": 0.04326663061114116,
|
|
"grad_norm": 2.703125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.5045,
|
|
"mean_token_accuracy": 0.039873688109219076,
|
|
"num_tokens": 88871.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.554053497314452,
|
|
"epoch": 0.048674959437533805,
|
|
"grad_norm": 2.71875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.509,
|
|
"mean_token_accuracy": 0.0394243435934186,
|
|
"num_tokens": 100118.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.538364601135253,
|
|
"epoch": 0.05408328826392644,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.4423,
|
|
"mean_token_accuracy": 0.048855461180210114,
|
|
"num_tokens": 113120.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.51687936782837,
|
|
"epoch": 0.05949161709031909,
|
|
"grad_norm": 2.6875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.3524,
|
|
"mean_token_accuracy": 0.046342721208930016,
|
|
"num_tokens": 123220.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.486348628997803,
|
|
"epoch": 0.06489994591671173,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.3335,
|
|
"mean_token_accuracy": 0.057121139764785764,
|
|
"num_tokens": 134479.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.347621631622314,
|
|
"epoch": 0.07030827474310439,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.1695,
|
|
"mean_token_accuracy": 0.06469983011484146,
|
|
"num_tokens": 145189.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.380140495300292,
|
|
"epoch": 0.07571660356949703,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.0849,
|
|
"mean_token_accuracy": 0.0679698098450899,
|
|
"num_tokens": 157354.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.407375621795655,
|
|
"epoch": 0.08112493239588967,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.0362,
|
|
"mean_token_accuracy": 0.06854141913354397,
|
|
"num_tokens": 168821.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.368484210968017,
|
|
"epoch": 0.08653326122228232,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 8.8889,
|
|
"mean_token_accuracy": 0.06982190646231175,
|
|
"num_tokens": 180554.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.238024425506591,
|
|
"epoch": 0.09194159004867496,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 8.7914,
|
|
"mean_token_accuracy": 0.07617409080266953,
|
|
"num_tokens": 191229.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.228119373321533,
|
|
"epoch": 0.09734991887506761,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.6865,
|
|
"mean_token_accuracy": 0.07864613682031632,
|
|
"num_tokens": 201903.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.117883205413818,
|
|
"epoch": 0.10275824770146025,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.6671,
|
|
"mean_token_accuracy": 0.07588070593774318,
|
|
"num_tokens": 214255.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.088370513916015,
|
|
"epoch": 0.10816657652785289,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.4926,
|
|
"mean_token_accuracy": 0.08170793950557709,
|
|
"num_tokens": 225423.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.01626272201538,
|
|
"epoch": 0.11357490535424554,
|
|
"grad_norm": 2.765625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.4216,
|
|
"mean_token_accuracy": 0.0811295285820961,
|
|
"num_tokens": 237801.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 9.888921546936036,
|
|
"epoch": 0.11898323418063818,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.2161,
|
|
"mean_token_accuracy": 0.08590030111372471,
|
|
"num_tokens": 249195.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.866813278198242,
|
|
"epoch": 0.12439156300703083,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.0932,
|
|
"mean_token_accuracy": 0.09252363666892052,
|
|
"num_tokens": 260129.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.781723117828369,
|
|
"epoch": 0.12979989183342347,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.1482,
|
|
"mean_token_accuracy": 0.07961882352828979,
|
|
"num_tokens": 271220.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.653579044342042,
|
|
"epoch": 0.1352082206598161,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.9073,
|
|
"mean_token_accuracy": 0.08841332867741585,
|
|
"num_tokens": 281560.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.472657203674316,
|
|
"epoch": 0.14061654948620878,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.7576,
|
|
"mean_token_accuracy": 0.08678923547267914,
|
|
"num_tokens": 292518.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.37501802444458,
|
|
"epoch": 0.1460248783126014,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.7439,
|
|
"mean_token_accuracy": 0.08266892358660698,
|
|
"num_tokens": 304865.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.135330295562744,
|
|
"epoch": 0.15143320713899405,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.5576,
|
|
"mean_token_accuracy": 0.0914510041475296,
|
|
"num_tokens": 315951.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.899916648864746,
|
|
"epoch": 0.1568415359653867,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.4698,
|
|
"mean_token_accuracy": 0.0939729243516922,
|
|
"num_tokens": 327094.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.730432033538818,
|
|
"epoch": 0.16224986479177933,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.4175,
|
|
"mean_token_accuracy": 0.09195762872695923,
|
|
"num_tokens": 337801.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.471641349792481,
|
|
"epoch": 0.167658193618172,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.3256,
|
|
"mean_token_accuracy": 0.09636828005313873,
|
|
"num_tokens": 349039.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.36181583404541,
|
|
"epoch": 0.17306652244456464,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.3141,
|
|
"mean_token_accuracy": 0.09097177907824516,
|
|
"num_tokens": 360639.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.211484432220459,
|
|
"epoch": 0.17847485127095727,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.3216,
|
|
"mean_token_accuracy": 0.09984223693609237,
|
|
"num_tokens": 371005.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.06982226371765,
|
|
"epoch": 0.1838831800973499,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.1972,
|
|
"mean_token_accuracy": 0.09530351907014847,
|
|
"num_tokens": 383340.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.89442892074585,
|
|
"epoch": 0.18929150892374255,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.1349,
|
|
"mean_token_accuracy": 0.10384258776903152,
|
|
"num_tokens": 394682.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.894894552230835,
|
|
"epoch": 0.19469983775013522,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.1506,
|
|
"mean_token_accuracy": 0.09678633958101272,
|
|
"num_tokens": 405713.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.920239067077636,
|
|
"epoch": 0.20010816657652786,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.0992,
|
|
"mean_token_accuracy": 0.09813495352864265,
|
|
"num_tokens": 416208.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.751987409591675,
|
|
"epoch": 0.2055164954029205,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.0093,
|
|
"mean_token_accuracy": 0.0977411150932312,
|
|
"num_tokens": 427595.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.748253011703492,
|
|
"epoch": 0.21092482422931313,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.1353,
|
|
"mean_token_accuracy": 0.10629882290959358,
|
|
"num_tokens": 438273.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.726998567581177,
|
|
"epoch": 0.21633315305570577,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.0456,
|
|
"mean_token_accuracy": 0.09703748449683189,
|
|
"num_tokens": 448948.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.69028058052063,
|
|
"epoch": 0.22174148188209844,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.0949,
|
|
"mean_token_accuracy": 0.09859743788838386,
|
|
"num_tokens": 460838.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.614819478988648,
|
|
"epoch": 0.22714981070849108,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.0979,
|
|
"mean_token_accuracy": 0.09938543438911437,
|
|
"num_tokens": 473555.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.5858986377716064,
|
|
"epoch": 0.23255813953488372,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.0348,
|
|
"mean_token_accuracy": 0.10209831669926643,
|
|
"num_tokens": 485221.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.542715787887573,
|
|
"epoch": 0.23796646836127636,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 6.9358,
|
|
"mean_token_accuracy": 0.10402886867523194,
|
|
"num_tokens": 496552.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.515826463699341,
|
|
"epoch": 0.24337479718766902,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.9993,
|
|
"mean_token_accuracy": 0.10345738977193833,
|
|
"num_tokens": 509760.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.480216789245605,
|
|
"epoch": 0.24878312601406166,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.8336,
|
|
"mean_token_accuracy": 0.11249553188681602,
|
|
"num_tokens": 521175.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.457189130783081,
|
|
"epoch": 0.2541914548404543,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.9476,
|
|
"mean_token_accuracy": 0.1091655321419239,
|
|
"num_tokens": 532597.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.4506001472473145,
|
|
"epoch": 0.25959978366684694,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.0133,
|
|
"mean_token_accuracy": 0.10449624136090278,
|
|
"num_tokens": 545661.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.514426422119141,
|
|
"epoch": 0.2650081124932396,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.979,
|
|
"mean_token_accuracy": 0.10253533497452735,
|
|
"num_tokens": 557336.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.388407945632935,
|
|
"epoch": 0.2704164413196322,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.8425,
|
|
"mean_token_accuracy": 0.11243968978524208,
|
|
"num_tokens": 567398.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.362952089309692,
|
|
"epoch": 0.27582477014602486,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.9312,
|
|
"mean_token_accuracy": 0.10559455379843712,
|
|
"num_tokens": 579543.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.340886926651001,
|
|
"epoch": 0.28123309897241755,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.753,
|
|
"mean_token_accuracy": 0.11401380077004433,
|
|
"num_tokens": 589218.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.349085855484009,
|
|
"epoch": 0.2866414277988102,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.8123,
|
|
"mean_token_accuracy": 0.11433196514844894,
|
|
"num_tokens": 600327.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.245560216903686,
|
|
"epoch": 0.2920497566252028,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.7717,
|
|
"mean_token_accuracy": 0.10991677045822143,
|
|
"num_tokens": 610923.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.36246280670166,
|
|
"epoch": 0.29745808545159547,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.8509,
|
|
"mean_token_accuracy": 0.10712052285671234,
|
|
"num_tokens": 621138.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.193227529525757,
|
|
"epoch": 0.3028664142779881,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.797,
|
|
"mean_token_accuracy": 0.11277740374207497,
|
|
"num_tokens": 631440.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.308440208435059,
|
|
"epoch": 0.30827474310438074,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.8392,
|
|
"mean_token_accuracy": 0.11829005107283593,
|
|
"num_tokens": 642328.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.2388709545135494,
|
|
"epoch": 0.3136830719307734,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.7809,
|
|
"mean_token_accuracy": 0.11447241306304931,
|
|
"num_tokens": 652875.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.163767290115357,
|
|
"epoch": 0.319091400757166,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.717,
|
|
"mean_token_accuracy": 0.1139206700026989,
|
|
"num_tokens": 663726.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.29945330619812,
|
|
"epoch": 0.32449972958355866,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.9178,
|
|
"mean_token_accuracy": 0.11128493323922158,
|
|
"num_tokens": 675638.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.235377931594849,
|
|
"epoch": 0.3299080584099513,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.7798,
|
|
"mean_token_accuracy": 0.11265585124492646,
|
|
"num_tokens": 687239.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.244372606277466,
|
|
"epoch": 0.335316387236344,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.8244,
|
|
"mean_token_accuracy": 0.10671208873391151,
|
|
"num_tokens": 698905.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.157739448547363,
|
|
"epoch": 0.34072471606273663,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.8115,
|
|
"mean_token_accuracy": 0.11196564212441444,
|
|
"num_tokens": 710602.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.112873458862305,
|
|
"epoch": 0.34613304488912927,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.7691,
|
|
"mean_token_accuracy": 0.11123333275318145,
|
|
"num_tokens": 721326.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.238367700576783,
|
|
"epoch": 0.3515413737155219,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.743,
|
|
"mean_token_accuracy": 0.11625269651412964,
|
|
"num_tokens": 732972.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.11000337600708,
|
|
"epoch": 0.35694970254191455,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.7174,
|
|
"mean_token_accuracy": 0.11805651634931565,
|
|
"num_tokens": 743624.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.168763875961304,
|
|
"epoch": 0.3623580313683072,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.8101,
|
|
"mean_token_accuracy": 0.11515277996659279,
|
|
"num_tokens": 756384.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.154894590377808,
|
|
"epoch": 0.3677663601946998,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.7909,
|
|
"mean_token_accuracy": 0.11520344465970993,
|
|
"num_tokens": 767256.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.1156007766723635,
|
|
"epoch": 0.37317468902109246,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.7043,
|
|
"mean_token_accuracy": 0.11901735439896584,
|
|
"num_tokens": 779059.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.027617883682251,
|
|
"epoch": 0.3785830178474851,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.8016,
|
|
"mean_token_accuracy": 0.11899174749851227,
|
|
"num_tokens": 790291.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.1015932083129885,
|
|
"epoch": 0.3839913466738778,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.7063,
|
|
"mean_token_accuracy": 0.11784811541438103,
|
|
"num_tokens": 801450.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.017753171920776,
|
|
"epoch": 0.38939967550027044,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.625,
|
|
"mean_token_accuracy": 0.1219206415116787,
|
|
"num_tokens": 812428.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.003644132614136,
|
|
"epoch": 0.3948080043266631,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.7244,
|
|
"mean_token_accuracy": 0.1133291907608509,
|
|
"num_tokens": 823644.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.151744842529297,
|
|
"epoch": 0.4002163331530557,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.7116,
|
|
"mean_token_accuracy": 0.12677086964249612,
|
|
"num_tokens": 834064.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.033484983444214,
|
|
"epoch": 0.40562466197944835,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.6546,
|
|
"mean_token_accuracy": 0.11925762817263603,
|
|
"num_tokens": 845207.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 6.970419311523438,
|
|
"epoch": 0.411032990805841,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.7181,
|
|
"mean_token_accuracy": 0.11609353423118592,
|
|
"num_tokens": 855564.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.12083477973938,
|
|
"epoch": 0.41644131963223363,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.7377,
|
|
"mean_token_accuracy": 0.12403716966509819,
|
|
"num_tokens": 866959.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 6.928013324737549,
|
|
"epoch": 0.42184964845862627,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.6585,
|
|
"mean_token_accuracy": 0.125290846824646,
|
|
"num_tokens": 877319.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.049751853942871,
|
|
"epoch": 0.4272579772850189,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.7489,
|
|
"mean_token_accuracy": 0.11747724786400796,
|
|
"num_tokens": 888037.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 6.912725925445557,
|
|
"epoch": 0.43266630611141155,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.4906,
|
|
"mean_token_accuracy": 0.12297609969973564,
|
|
"num_tokens": 898775.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.0705005645751955,
|
|
"epoch": 0.43807463493780424,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.6433,
|
|
"mean_token_accuracy": 0.12098436206579208,
|
|
"num_tokens": 909513.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 6.938232803344727,
|
|
"epoch": 0.4434829637641969,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.6959,
|
|
"mean_token_accuracy": 0.11758999228477478,
|
|
"num_tokens": 920730.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 6.924176979064941,
|
|
"epoch": 0.4488912925905895,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.6151,
|
|
"mean_token_accuracy": 0.12237995192408561,
|
|
"num_tokens": 931457.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 6.865782737731934,
|
|
"epoch": 0.45429962141698216,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.6026,
|
|
"mean_token_accuracy": 0.12269196733832359,
|
|
"num_tokens": 942572.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 6.9836249351501465,
|
|
"epoch": 0.4597079502433748,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.607,
|
|
"mean_token_accuracy": 0.11696747690439224,
|
|
"num_tokens": 953132.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 6.915006303787232,
|
|
"epoch": 0.46511627906976744,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.6414,
|
|
"mean_token_accuracy": 0.11870383992791175,
|
|
"num_tokens": 963551.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 6.957869148254394,
|
|
"epoch": 0.4705246078961601,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.599,
|
|
"mean_token_accuracy": 0.12203558534383774,
|
|
"num_tokens": 975297.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.861082458496094,
|
|
"epoch": 0.4759329367225527,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.6076,
|
|
"mean_token_accuracy": 0.12166279256343841,
|
|
"num_tokens": 986910.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 6.963198518753051,
|
|
"epoch": 0.48134126554894535,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.7303,
|
|
"mean_token_accuracy": 0.12663792818784714,
|
|
"num_tokens": 997058.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.887674808502197,
|
|
"epoch": 0.48674959437533805,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.6118,
|
|
"mean_token_accuracy": 0.11577775105834007,
|
|
"num_tokens": 1008261.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.85930438041687,
|
|
"epoch": 0.4921579232017307,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.527,
|
|
"mean_token_accuracy": 0.1269462674856186,
|
|
"num_tokens": 1019193.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.86927375793457,
|
|
"epoch": 0.4975662520281233,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.6448,
|
|
"mean_token_accuracy": 0.12487371489405633,
|
|
"num_tokens": 1031227.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.828432273864746,
|
|
"epoch": 0.502974580854516,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.4702,
|
|
"mean_token_accuracy": 0.1266406588256359,
|
|
"num_tokens": 1042963.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.750533151626587,
|
|
"epoch": 0.5083829096809086,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.5397,
|
|
"mean_token_accuracy": 0.11756750047206879,
|
|
"num_tokens": 1053709.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.890934610366822,
|
|
"epoch": 0.5137912385073012,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.5681,
|
|
"mean_token_accuracy": 0.13536556139588357,
|
|
"num_tokens": 1064527.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.778618907928466,
|
|
"epoch": 0.5191995673336939,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.6317,
|
|
"mean_token_accuracy": 0.12472611665725708,
|
|
"num_tokens": 1075127.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.814949655532837,
|
|
"epoch": 0.5246078961600865,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.6266,
|
|
"mean_token_accuracy": 0.11958301439881325,
|
|
"num_tokens": 1086581.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.7194966793060305,
|
|
"epoch": 0.5300162249864792,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.55,
|
|
"mean_token_accuracy": 0.13253298550844192,
|
|
"num_tokens": 1097818.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.829859018325806,
|
|
"epoch": 0.5354245538128718,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.5252,
|
|
"mean_token_accuracy": 0.12295517474412918,
|
|
"num_tokens": 1108043.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.82882833480835,
|
|
"epoch": 0.5408328826392644,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.592,
|
|
"mean_token_accuracy": 0.12201916500926017,
|
|
"num_tokens": 1119875.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.7380537509918215,
|
|
"epoch": 0.5462412114656571,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.4926,
|
|
"mean_token_accuracy": 0.1267081081867218,
|
|
"num_tokens": 1132406.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.946386671066284,
|
|
"epoch": 0.5516495402920497,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.6676,
|
|
"mean_token_accuracy": 0.11859372109174729,
|
|
"num_tokens": 1143828.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.720641136169434,
|
|
"epoch": 0.5570578691184424,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.4863,
|
|
"mean_token_accuracy": 0.12196403294801712,
|
|
"num_tokens": 1155482.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.651684331893921,
|
|
"epoch": 0.5624661979448351,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.4324,
|
|
"mean_token_accuracy": 0.13508962318301201,
|
|
"num_tokens": 1166847.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.7738018989562985,
|
|
"epoch": 0.5678745267712277,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.5108,
|
|
"mean_token_accuracy": 0.1282059669494629,
|
|
"num_tokens": 1177419.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.822676420211792,
|
|
"epoch": 0.5732828555976204,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.6815,
|
|
"mean_token_accuracy": 0.11861670166254043,
|
|
"num_tokens": 1188975.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.75103759765625,
|
|
"epoch": 0.578691184424013,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.5153,
|
|
"mean_token_accuracy": 0.12072157859802246,
|
|
"num_tokens": 1200252.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.676786279678344,
|
|
"epoch": 0.5840995132504057,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.5459,
|
|
"mean_token_accuracy": 0.13504153192043306,
|
|
"num_tokens": 1211864.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.7417017936706545,
|
|
"epoch": 0.5895078420767983,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.4194,
|
|
"mean_token_accuracy": 0.12843444123864173,
|
|
"num_tokens": 1222429.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.745041131973267,
|
|
"epoch": 0.5949161709031909,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.5458,
|
|
"mean_token_accuracy": 0.12370594069361687,
|
|
"num_tokens": 1233782.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.790885591506958,
|
|
"epoch": 0.6003244997295836,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.4923,
|
|
"mean_token_accuracy": 0.12668371945619583,
|
|
"num_tokens": 1244309.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.6354663372039795,
|
|
"epoch": 0.6057328285559762,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.4497,
|
|
"mean_token_accuracy": 0.12727595791220664,
|
|
"num_tokens": 1256403.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.775501108169555,
|
|
"epoch": 0.6111411573823688,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.5459,
|
|
"mean_token_accuracy": 0.1229689784348011,
|
|
"num_tokens": 1268758.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.60977520942688,
|
|
"epoch": 0.6165494862087615,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.4351,
|
|
"mean_token_accuracy": 0.1309477686882019,
|
|
"num_tokens": 1279355.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.6845152378082275,
|
|
"epoch": 0.6219578150351541,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.4289,
|
|
"mean_token_accuracy": 0.12401615604758262,
|
|
"num_tokens": 1290619.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.649652814865112,
|
|
"epoch": 0.6273661438615468,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.3938,
|
|
"mean_token_accuracy": 0.13154794424772262,
|
|
"num_tokens": 1301851.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.745173740386963,
|
|
"epoch": 0.6327744726879394,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.5466,
|
|
"mean_token_accuracy": 0.13009120672941207,
|
|
"num_tokens": 1313260.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.597011137008667,
|
|
"epoch": 0.638182801514332,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.5331,
|
|
"mean_token_accuracy": 0.12869492620229722,
|
|
"num_tokens": 1324189.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.752423334121704,
|
|
"epoch": 0.6435911303407247,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.5636,
|
|
"mean_token_accuracy": 0.12625337839126588,
|
|
"num_tokens": 1335765.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.725863218307495,
|
|
"epoch": 0.6489994591671173,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.4908,
|
|
"mean_token_accuracy": 0.12820648178458213,
|
|
"num_tokens": 1347653.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.657638025283814,
|
|
"epoch": 0.65440778799351,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.4944,
|
|
"mean_token_accuracy": 0.12864069044589996,
|
|
"num_tokens": 1360201.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.614993000030518,
|
|
"epoch": 0.6598161168199026,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.354,
|
|
"mean_token_accuracy": 0.1303904466331005,
|
|
"num_tokens": 1370974.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.537722396850586,
|
|
"epoch": 0.6652244456462953,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.4304,
|
|
"mean_token_accuracy": 0.1266734465956688,
|
|
"num_tokens": 1383855.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.5735701560974125,
|
|
"epoch": 0.670632774472688,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.3969,
|
|
"mean_token_accuracy": 0.13598421812057496,
|
|
"num_tokens": 1395041.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.759116458892822,
|
|
"epoch": 0.6760411032990806,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.5041,
|
|
"mean_token_accuracy": 0.12941347509622575,
|
|
"num_tokens": 1406102.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.4543132305145265,
|
|
"epoch": 0.6814494321254733,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.3253,
|
|
"mean_token_accuracy": 0.12973413094878197,
|
|
"num_tokens": 1417713.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.5458276748657225,
|
|
"epoch": 0.6868577609518659,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.4712,
|
|
"mean_token_accuracy": 0.13443209007382392,
|
|
"num_tokens": 1428526.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.634454774856567,
|
|
"epoch": 0.6922660897782585,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.4175,
|
|
"mean_token_accuracy": 0.13237082809209824,
|
|
"num_tokens": 1438849.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.572102212905884,
|
|
"epoch": 0.6976744186046512,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.4857,
|
|
"mean_token_accuracy": 0.12762622609734536,
|
|
"num_tokens": 1450776.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.548601722717285,
|
|
"epoch": 0.7030827474310438,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.3406,
|
|
"mean_token_accuracy": 0.12978718280792237,
|
|
"num_tokens": 1462552.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.477000045776367,
|
|
"epoch": 0.7084910762574365,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.3594,
|
|
"mean_token_accuracy": 0.1336737334728241,
|
|
"num_tokens": 1473867.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.506887197494507,
|
|
"epoch": 0.7138994050838291,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.3755,
|
|
"mean_token_accuracy": 0.13364285305142404,
|
|
"num_tokens": 1485233.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.57369909286499,
|
|
"epoch": 0.7193077339102217,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.4229,
|
|
"mean_token_accuracy": 0.1384587913751602,
|
|
"num_tokens": 1496919.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.569575452804566,
|
|
"epoch": 0.7247160627366144,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.3744,
|
|
"mean_token_accuracy": 0.12899189442396164,
|
|
"num_tokens": 1508695.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.57951979637146,
|
|
"epoch": 0.730124391563007,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.4236,
|
|
"mean_token_accuracy": 0.13620161265134811,
|
|
"num_tokens": 1519629.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.527429914474487,
|
|
"epoch": 0.7355327203893997,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.3362,
|
|
"mean_token_accuracy": 0.13338930681347846,
|
|
"num_tokens": 1530144.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.445765209197998,
|
|
"epoch": 0.7409410492157923,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.338,
|
|
"mean_token_accuracy": 0.13429746255278588,
|
|
"num_tokens": 1541013.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.607530164718628,
|
|
"epoch": 0.7463493780421849,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.4094,
|
|
"mean_token_accuracy": 0.13534171059727668,
|
|
"num_tokens": 1553344.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.519038200378418,
|
|
"epoch": 0.7517577068685776,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.4489,
|
|
"mean_token_accuracy": 0.13384861946105958,
|
|
"num_tokens": 1565298.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.506717586517334,
|
|
"epoch": 0.7571660356949702,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.3857,
|
|
"mean_token_accuracy": 0.14263246059417725,
|
|
"num_tokens": 1577645.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.445629692077636,
|
|
"epoch": 0.7625743645213628,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.3658,
|
|
"mean_token_accuracy": 0.13934621810913086,
|
|
"num_tokens": 1588387.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.639528369903564,
|
|
"epoch": 0.7679826933477556,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.4506,
|
|
"mean_token_accuracy": 0.1386639840900898,
|
|
"num_tokens": 1599837.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.476478481292725,
|
|
"epoch": 0.7733910221741482,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.3896,
|
|
"mean_token_accuracy": 0.12935182079672813,
|
|
"num_tokens": 1610401.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.4148335456848145,
|
|
"epoch": 0.7787993510005409,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.3446,
|
|
"mean_token_accuracy": 0.13272809460759163,
|
|
"num_tokens": 1622688.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.624324560165405,
|
|
"epoch": 0.7842076798269335,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.4146,
|
|
"mean_token_accuracy": 0.12909135445952416,
|
|
"num_tokens": 1633594.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.3749419212341305,
|
|
"epoch": 0.7896160086533262,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.3181,
|
|
"mean_token_accuracy": 0.13193461969494819,
|
|
"num_tokens": 1645861.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.385309505462646,
|
|
"epoch": 0.7950243374797188,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.2236,
|
|
"mean_token_accuracy": 0.14419274926185607,
|
|
"num_tokens": 1656903.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.447410774230957,
|
|
"epoch": 0.8004326663061114,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.3334,
|
|
"mean_token_accuracy": 0.13758712783455848,
|
|
"num_tokens": 1668130.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.309524917602539,
|
|
"epoch": 0.8058409951325041,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.2358,
|
|
"mean_token_accuracy": 0.14352481439709663,
|
|
"num_tokens": 1678228.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.534902763366699,
|
|
"epoch": 0.8112493239588967,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.3822,
|
|
"mean_token_accuracy": 0.13744921386241912,
|
|
"num_tokens": 1690339.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.490798282623291,
|
|
"epoch": 0.8166576527852893,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.3928,
|
|
"mean_token_accuracy": 0.12897668555378913,
|
|
"num_tokens": 1701067.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.508419227600098,
|
|
"epoch": 0.822065981611682,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.3318,
|
|
"mean_token_accuracy": 0.14198972061276435,
|
|
"num_tokens": 1712835.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.4104822158813475,
|
|
"epoch": 0.8274743104380746,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.2808,
|
|
"mean_token_accuracy": 0.13734794408082962,
|
|
"num_tokens": 1723432.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.467193269729615,
|
|
"epoch": 0.8328826392644673,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.3242,
|
|
"mean_token_accuracy": 0.13243861123919487,
|
|
"num_tokens": 1734134.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.3534571647644045,
|
|
"epoch": 0.8382909680908599,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.2744,
|
|
"mean_token_accuracy": 0.1389221429824829,
|
|
"num_tokens": 1744723.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.501026391983032,
|
|
"epoch": 0.8436992969172525,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.4111,
|
|
"mean_token_accuracy": 0.13507046177983284,
|
|
"num_tokens": 1755269.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.521368360519409,
|
|
"epoch": 0.8491076257436452,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.4018,
|
|
"mean_token_accuracy": 0.1358663931488991,
|
|
"num_tokens": 1766344.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.426554012298584,
|
|
"epoch": 0.8545159545700378,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.3736,
|
|
"mean_token_accuracy": 0.12991632744669915,
|
|
"num_tokens": 1779052.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.335146522521972,
|
|
"epoch": 0.8599242833964305,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.2379,
|
|
"mean_token_accuracy": 0.13654442951083184,
|
|
"num_tokens": 1789754.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.265465784072876,
|
|
"epoch": 0.8653326122228231,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.1662,
|
|
"mean_token_accuracy": 0.14170663058757782,
|
|
"num_tokens": 1801161.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.439857339859008,
|
|
"epoch": 0.8707409410492158,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.2636,
|
|
"mean_token_accuracy": 0.13717492818832397,
|
|
"num_tokens": 1810892.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.293186140060425,
|
|
"epoch": 0.8761492698756085,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.3348,
|
|
"mean_token_accuracy": 0.134286767244339,
|
|
"num_tokens": 1821574.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.438879680633545,
|
|
"epoch": 0.8815575987020011,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.3697,
|
|
"mean_token_accuracy": 0.13711444959044455,
|
|
"num_tokens": 1833342.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.382972478866577,
|
|
"epoch": 0.8869659275283938,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.2438,
|
|
"mean_token_accuracy": 0.13821374326944352,
|
|
"num_tokens": 1843768.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.487565708160401,
|
|
"epoch": 0.8923742563547864,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.3156,
|
|
"mean_token_accuracy": 0.13171282932162284,
|
|
"num_tokens": 1855569.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.3501959323883055,
|
|
"epoch": 0.897782585181179,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.3051,
|
|
"mean_token_accuracy": 0.13134763911366462,
|
|
"num_tokens": 1866684.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.362168693542481,
|
|
"epoch": 0.9031909140075717,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.2304,
|
|
"mean_token_accuracy": 0.13782247975468637,
|
|
"num_tokens": 1878700.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.378588628768921,
|
|
"epoch": 0.9085992428339643,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.2563,
|
|
"mean_token_accuracy": 0.13868874236941336,
|
|
"num_tokens": 1889970.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.256662464141845,
|
|
"epoch": 0.914007571660357,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.1686,
|
|
"mean_token_accuracy": 0.13953565880656243,
|
|
"num_tokens": 1902008.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.318142700195312,
|
|
"epoch": 0.9194159004867496,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.3913,
|
|
"mean_token_accuracy": 0.13987977653741837,
|
|
"num_tokens": 1913253.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.398606395721435,
|
|
"epoch": 0.9248242293131422,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.242,
|
|
"mean_token_accuracy": 0.14221547394990922,
|
|
"num_tokens": 1923408.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.3646495819091795,
|
|
"epoch": 0.9302325581395349,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.299,
|
|
"mean_token_accuracy": 0.1426120236515999,
|
|
"num_tokens": 1935315.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.220848560333252,
|
|
"epoch": 0.9356408869659275,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.1476,
|
|
"mean_token_accuracy": 0.14092365950345992,
|
|
"num_tokens": 1946524.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.3639672756195065,
|
|
"epoch": 0.9410492157923201,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.2897,
|
|
"mean_token_accuracy": 0.13832299113273622,
|
|
"num_tokens": 1957561.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.228976774215698,
|
|
"epoch": 0.9464575446187128,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.2271,
|
|
"mean_token_accuracy": 0.13985422402620315,
|
|
"num_tokens": 1969089.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.3985954284667965,
|
|
"epoch": 0.9518658734451054,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.2572,
|
|
"mean_token_accuracy": 0.13929264023900031,
|
|
"num_tokens": 1980341.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.194135761260986,
|
|
"epoch": 0.9572742022714981,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.2152,
|
|
"mean_token_accuracy": 0.1457873597741127,
|
|
"num_tokens": 1991220.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.379308748245239,
|
|
"epoch": 0.9626825310978907,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.306,
|
|
"mean_token_accuracy": 0.14043606221675872,
|
|
"num_tokens": 2002650.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.273330116271973,
|
|
"epoch": 0.9680908599242833,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.2088,
|
|
"mean_token_accuracy": 0.14419417828321457,
|
|
"num_tokens": 2012884.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.3799642562866214,
|
|
"epoch": 0.9734991887506761,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.2861,
|
|
"mean_token_accuracy": 0.13549073413014412,
|
|
"num_tokens": 2023543.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.387584972381592,
|
|
"epoch": 0.9789075175770687,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.3488,
|
|
"mean_token_accuracy": 0.13450675085186958,
|
|
"num_tokens": 2034789.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.239657592773438,
|
|
"epoch": 0.9843158464034614,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.2558,
|
|
"mean_token_accuracy": 0.1416017398238182,
|
|
"num_tokens": 2046719.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.252905702590942,
|
|
"epoch": 0.989724175229854,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.1984,
|
|
"mean_token_accuracy": 0.14273628294467927,
|
|
"num_tokens": 2057016.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.232679033279419,
|
|
"epoch": 0.9951325040562466,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.193,
|
|
"mean_token_accuracy": 0.1355368375778198,
|
|
"num_tokens": 2067898.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.277052084604899,
|
|
"epoch": 1.0,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.2098,
|
|
"mean_token_accuracy": 0.1415335453218884,
|
|
"num_tokens": 2076889.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.290318346023559,
|
|
"epoch": 1.0054083288263926,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.1099,
|
|
"mean_token_accuracy": 0.13955417200922965,
|
|
"num_tokens": 2089228.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.165477848052978,
|
|
"epoch": 1.0108166576527853,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.1142,
|
|
"mean_token_accuracy": 0.1436012014746666,
|
|
"num_tokens": 2101098.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.25781569480896,
|
|
"epoch": 1.016224986479178,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.0536,
|
|
"mean_token_accuracy": 0.1513492174446583,
|
|
"num_tokens": 2113081.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.2349457263946535,
|
|
"epoch": 1.0216333153055706,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.0386,
|
|
"mean_token_accuracy": 0.15058111846446992,
|
|
"num_tokens": 2123356.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.323075008392334,
|
|
"epoch": 1.0270416441319632,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.1321,
|
|
"mean_token_accuracy": 0.14740860909223558,
|
|
"num_tokens": 2134388.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.072897148132324,
|
|
"epoch": 1.0324499729583558,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.013,
|
|
"mean_token_accuracy": 0.15088868290185928,
|
|
"num_tokens": 2145209.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.271716213226318,
|
|
"epoch": 1.0378583017847485,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.0733,
|
|
"mean_token_accuracy": 0.15297747403383255,
|
|
"num_tokens": 2157146.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.113353109359741,
|
|
"epoch": 1.043266630611141,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.0832,
|
|
"mean_token_accuracy": 0.14374596923589705,
|
|
"num_tokens": 2168721.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.233334302902222,
|
|
"epoch": 1.0486749594375337,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.0774,
|
|
"mean_token_accuracy": 0.14971460700035094,
|
|
"num_tokens": 2179213.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.177529621124267,
|
|
"epoch": 1.0540832882639264,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.1019,
|
|
"mean_token_accuracy": 0.14426251277327537,
|
|
"num_tokens": 2191224.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.3130913257598875,
|
|
"epoch": 1.059491617090319,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.1694,
|
|
"mean_token_accuracy": 0.14404836148023606,
|
|
"num_tokens": 2203665.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.18894772529602,
|
|
"epoch": 1.0648999459167117,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.0744,
|
|
"mean_token_accuracy": 0.14550448656082154,
|
|
"num_tokens": 2215136.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.178538846969604,
|
|
"epoch": 1.0703082747431043,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.1213,
|
|
"mean_token_accuracy": 0.15047272816300392,
|
|
"num_tokens": 2226375.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.113742542266846,
|
|
"epoch": 1.075716603569497,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 5.967,
|
|
"mean_token_accuracy": 0.15593916028738022,
|
|
"num_tokens": 2236097.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.344080114364624,
|
|
"epoch": 1.0811249323958896,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.2159,
|
|
"mean_token_accuracy": 0.14157222136855124,
|
|
"num_tokens": 2247964.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0811249323958896,
|
|
"eval_entropy": 6.1291101856665176,
|
|
"eval_loss": 6.394670009613037,
|
|
"eval_mean_token_accuracy": 0.1441675969907506,
|
|
"eval_num_tokens": 2247964.0,
|
|
"eval_runtime": 2.5419,
|
|
"eval_samples_per_second": 1382.851,
|
|
"eval_steps_per_second": 173.102,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.142773246765136,
|
|
"epoch": 1.0865332612222822,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999997383518051,
|
|
"loss": 6.1585,
|
|
"mean_token_accuracy": 0.15236814618110656,
|
|
"num_tokens": 2260467.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.1961956977844235,
|
|
"epoch": 1.0919415900486749,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999986754070562,
|
|
"loss": 6.0693,
|
|
"mean_token_accuracy": 0.14688059836626052,
|
|
"num_tokens": 2272176.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.236271953582763,
|
|
"epoch": 1.0973499188750675,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004999967948166009,
|
|
"loss": 6.1149,
|
|
"mean_token_accuracy": 0.14699096828699112,
|
|
"num_tokens": 2283491.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.119901418685913,
|
|
"epoch": 1.1027582477014604,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999940965872734,
|
|
"loss": 6.0037,
|
|
"mean_token_accuracy": 0.14384915679693222,
|
|
"num_tokens": 2295131.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.100546836853027,
|
|
"epoch": 1.1081665765278528,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999905807288788,
|
|
"loss": 6.0291,
|
|
"mean_token_accuracy": 0.1478397861123085,
|
|
"num_tokens": 2305832.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.050093078613282,
|
|
"epoch": 1.1135749053542456,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000499986247254194,
|
|
"loss": 6.0281,
|
|
"mean_token_accuracy": 0.15077048987150193,
|
|
"num_tokens": 2315874.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.332290458679199,
|
|
"epoch": 1.1189832341806383,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999810961789667,
|
|
"loss": 6.2145,
|
|
"mean_token_accuracy": 0.14445521235466002,
|
|
"num_tokens": 2327580.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.1578165054321286,
|
|
"epoch": 1.124391563007031,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499975127521916,
|
|
"loss": 6.0434,
|
|
"mean_token_accuracy": 0.14994928389787673,
|
|
"num_tokens": 2340426.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.16860818862915,
|
|
"epoch": 1.1297998918334236,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999683413047319,
|
|
"loss": 6.1767,
|
|
"mean_token_accuracy": 0.14190550222992898,
|
|
"num_tokens": 2353228.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.170330619812011,
|
|
"epoch": 1.1352082206598162,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999607375520759,
|
|
"loss": 5.9928,
|
|
"mean_token_accuracy": 0.15112272799015045,
|
|
"num_tokens": 2364238.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.134090328216553,
|
|
"epoch": 1.1406165494862088,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999523162915794,
|
|
"loss": 6.1344,
|
|
"mean_token_accuracy": 0.14725330024957656,
|
|
"num_tokens": 2375847.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.104169321060181,
|
|
"epoch": 1.1460248783126015,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999430775538459,
|
|
"loss": 6.0251,
|
|
"mean_token_accuracy": 0.15378995686769487,
|
|
"num_tokens": 2386489.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.1245194435119625,
|
|
"epoch": 1.151433207138994,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999330213724487,
|
|
"loss": 6.0558,
|
|
"mean_token_accuracy": 0.14770212918519973,
|
|
"num_tokens": 2398839.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.17924132347107,
|
|
"epoch": 1.1568415359653867,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999221477839319,
|
|
"loss": 6.0275,
|
|
"mean_token_accuracy": 0.15129323601722716,
|
|
"num_tokens": 2410016.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.092270135879517,
|
|
"epoch": 1.1622498647917794,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999104568278104,
|
|
"loss": 6.0562,
|
|
"mean_token_accuracy": 0.15010673999786378,
|
|
"num_tokens": 2421202.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.230505514144897,
|
|
"epoch": 1.167658193618172,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499897948546569,
|
|
"loss": 6.1418,
|
|
"mean_token_accuracy": 0.1431541547179222,
|
|
"num_tokens": 2431943.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.030719041824341,
|
|
"epoch": 1.1730665224445647,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998846229856629,
|
|
"loss": 5.961,
|
|
"mean_token_accuracy": 0.14953965842723846,
|
|
"num_tokens": 2443454.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.230849647521973,
|
|
"epoch": 1.1784748512709573,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998704801935171,
|
|
"loss": 6.1651,
|
|
"mean_token_accuracy": 0.14004991054534913,
|
|
"num_tokens": 2455909.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.0102503299713135,
|
|
"epoch": 1.18388318009735,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998555202215267,
|
|
"loss": 6.0293,
|
|
"mean_token_accuracy": 0.152455472946167,
|
|
"num_tokens": 2467486.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.07820725440979,
|
|
"epoch": 1.1892915089237426,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998397431240561,
|
|
"loss": 5.9762,
|
|
"mean_token_accuracy": 0.15628656297922133,
|
|
"num_tokens": 2477560.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.092442607879638,
|
|
"epoch": 1.1946998377501352,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998231489584395,
|
|
"loss": 6.0363,
|
|
"mean_token_accuracy": 0.15760239511728286,
|
|
"num_tokens": 2487222.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.095543098449707,
|
|
"epoch": 1.2001081665765279,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00049980573778498,
|
|
"loss": 6.0451,
|
|
"mean_token_accuracy": 0.1515233427286148,
|
|
"num_tokens": 2497474.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.1058553695678714,
|
|
"epoch": 1.2055164954029205,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997875096669498,
|
|
"loss": 6.1091,
|
|
"mean_token_accuracy": 0.14928437322378157,
|
|
"num_tokens": 2508185.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.161521863937378,
|
|
"epoch": 1.2109248242293131,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997684646705901,
|
|
"loss": 6.0486,
|
|
"mean_token_accuracy": 0.1468924105167389,
|
|
"num_tokens": 2518876.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.0585887908935545,
|
|
"epoch": 1.2163331530557058,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997486028651103,
|
|
"loss": 6.0454,
|
|
"mean_token_accuracy": 0.15654876828193665,
|
|
"num_tokens": 2530554.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 5.991144180297852,
|
|
"epoch": 1.2217414818820984,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004997279243226884,
|
|
"loss": 5.9299,
|
|
"mean_token_accuracy": 0.15320944488048555,
|
|
"num_tokens": 2541331.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.175770854949951,
|
|
"epoch": 1.227149810708491,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997064291184701,
|
|
"loss": 6.1457,
|
|
"mean_token_accuracy": 0.1422166146337986,
|
|
"num_tokens": 2554225.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 5.990706539154052,
|
|
"epoch": 1.2325581395348837,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996841173305691,
|
|
"loss": 6.0335,
|
|
"mean_token_accuracy": 0.15032867193222046,
|
|
"num_tokens": 2565050.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.201693868637085,
|
|
"epoch": 1.2379664683612763,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996609890400665,
|
|
"loss": 6.1237,
|
|
"mean_token_accuracy": 0.14705591276288033,
|
|
"num_tokens": 2576926.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.258083057403565,
|
|
"epoch": 1.243374797187669,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004996370443310107,
|
|
"loss": 6.1839,
|
|
"mean_token_accuracy": 0.14229817315936089,
|
|
"num_tokens": 2588775.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 5.996690702438355,
|
|
"epoch": 1.2487831260140616,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004996122832904165,
|
|
"loss": 6.0329,
|
|
"mean_token_accuracy": 0.15009836107492447,
|
|
"num_tokens": 2600534.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.008330345153809,
|
|
"epoch": 1.2541914548404542,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995867060082658,
|
|
"loss": 5.8658,
|
|
"mean_token_accuracy": 0.1588037133216858,
|
|
"num_tokens": 2612350.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.040723133087158,
|
|
"epoch": 1.2595997836668469,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004995603125775065,
|
|
"loss": 6.0462,
|
|
"mean_token_accuracy": 0.14463529512286186,
|
|
"num_tokens": 2623871.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.083625268936157,
|
|
"epoch": 1.2650081124932395,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004995331030940523,
|
|
"loss": 5.9817,
|
|
"mean_token_accuracy": 0.15036386251449585,
|
|
"num_tokens": 2636371.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.1035080909729,
|
|
"epoch": 1.2704164413196322,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004995050776567826,
|
|
"loss": 6.0369,
|
|
"mean_token_accuracy": 0.1542038917541504,
|
|
"num_tokens": 2646794.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.0394493579864506,
|
|
"epoch": 1.2758247701460248,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004994762363675421,
|
|
"loss": 6.0347,
|
|
"mean_token_accuracy": 0.15617663115262986,
|
|
"num_tokens": 2657105.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.043619441986084,
|
|
"epoch": 1.2812330989724177,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004994465793311396,
|
|
"loss": 5.9675,
|
|
"mean_token_accuracy": 0.1518074244260788,
|
|
"num_tokens": 2668297.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.004485082626343,
|
|
"epoch": 1.28664142779881,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004994161066553493,
|
|
"loss": 5.8288,
|
|
"mean_token_accuracy": 0.15841025859117508,
|
|
"num_tokens": 2678834.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.030247688293457,
|
|
"epoch": 1.292049756625203,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004993848184509086,
|
|
"loss": 6.0713,
|
|
"mean_token_accuracy": 0.15047259926795958,
|
|
"num_tokens": 2690392.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 5.898325634002686,
|
|
"epoch": 1.2974580854515954,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004993527148315193,
|
|
"loss": 5.8678,
|
|
"mean_token_accuracy": 0.15419968068599701,
|
|
"num_tokens": 2701319.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.144197940826416,
|
|
"epoch": 1.3028664142779882,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004993197959138458,
|
|
"loss": 6.0435,
|
|
"mean_token_accuracy": 0.14768160730600358,
|
|
"num_tokens": 2713018.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 5.988416862487793,
|
|
"epoch": 1.3082747431043806,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004992860618175156,
|
|
"loss": 6.1553,
|
|
"mean_token_accuracy": 0.14605925753712654,
|
|
"num_tokens": 2724995.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.152584838867187,
|
|
"epoch": 1.3136830719307735,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004992515126651184,
|
|
"loss": 5.9738,
|
|
"mean_token_accuracy": 0.153250615298748,
|
|
"num_tokens": 2736733.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 5.910708856582642,
|
|
"epoch": 1.319091400757166,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004992161485822059,
|
|
"loss": 5.9658,
|
|
"mean_token_accuracy": 0.15414738059043884,
|
|
"num_tokens": 2748436.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 5.983804845809937,
|
|
"epoch": 1.3244997295835588,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004991799696972913,
|
|
"loss": 5.9864,
|
|
"mean_token_accuracy": 0.15305898189544678,
|
|
"num_tokens": 2758924.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.143850517272949,
|
|
"epoch": 1.3299080584099512,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004991429761418488,
|
|
"loss": 6.0278,
|
|
"mean_token_accuracy": 0.14822815358638763,
|
|
"num_tokens": 2769773.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.167329740524292,
|
|
"epoch": 1.335316387236344,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000499105168050313,
|
|
"loss": 6.1744,
|
|
"mean_token_accuracy": 0.15468801707029342,
|
|
"num_tokens": 2782414.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 5.907335376739502,
|
|
"epoch": 1.3407247160627367,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004990665455600787,
|
|
"loss": 5.9338,
|
|
"mean_token_accuracy": 0.15370485186576843,
|
|
"num_tokens": 2794325.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 5.907944393157959,
|
|
"epoch": 1.3461330448891293,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004990271088114999,
|
|
"loss": 5.7784,
|
|
"mean_token_accuracy": 0.16055067330598832,
|
|
"num_tokens": 2805391.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 5.9189352035522464,
|
|
"epoch": 1.351541373715522,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004989868579478901,
|
|
"loss": 5.9042,
|
|
"mean_token_accuracy": 0.15601754039525986,
|
|
"num_tokens": 2816633.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.080076599121094,
|
|
"epoch": 1.3569497025419146,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004989457931155209,
|
|
"loss": 6.0248,
|
|
"mean_token_accuracy": 0.15658325403928758,
|
|
"num_tokens": 2828395.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.081086015701294,
|
|
"epoch": 1.3623580313683072,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004989039144636219,
|
|
"loss": 6.0136,
|
|
"mean_token_accuracy": 0.15574034601449965,
|
|
"num_tokens": 2839592.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 5.983651113510132,
|
|
"epoch": 1.3677663601946999,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004988612221443804,
|
|
"loss": 5.9822,
|
|
"mean_token_accuracy": 0.16441779136657714,
|
|
"num_tokens": 2849557.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.097548580169677,
|
|
"epoch": 1.3731746890210925,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004988177163129403,
|
|
"loss": 6.0325,
|
|
"mean_token_accuracy": 0.15445336997509002,
|
|
"num_tokens": 2862030.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 5.900971508026123,
|
|
"epoch": 1.3785830178474852,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004987733971274018,
|
|
"loss": 5.9669,
|
|
"mean_token_accuracy": 0.15942161679267883,
|
|
"num_tokens": 2872379.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.088207578659057,
|
|
"epoch": 1.3839913466738778,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004987282647488211,
|
|
"loss": 6.0436,
|
|
"mean_token_accuracy": 0.1526199296116829,
|
|
"num_tokens": 2883170.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 5.986858177185058,
|
|
"epoch": 1.3893996755002704,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004986823193412093,
|
|
"loss": 5.9588,
|
|
"mean_token_accuracy": 0.15029667615890502,
|
|
"num_tokens": 2894013.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.065160417556763,
|
|
"epoch": 1.394808004326663,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000498635561071532,
|
|
"loss": 6.0517,
|
|
"mean_token_accuracy": 0.15685324519872665,
|
|
"num_tokens": 2905073.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.072223758697509,
|
|
"epoch": 1.4002163331530557,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000498587990109709,
|
|
"loss": 6.0164,
|
|
"mean_token_accuracy": 0.15148179829120637,
|
|
"num_tokens": 2916579.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 5.9010703563690186,
|
|
"epoch": 1.4056246619794484,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004985396066286133,
|
|
"loss": 5.9092,
|
|
"mean_token_accuracy": 0.15624812096357346,
|
|
"num_tokens": 2928522.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.052590990066529,
|
|
"epoch": 1.411032990805841,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004984904108040706,
|
|
"loss": 5.996,
|
|
"mean_token_accuracy": 0.15711974650621413,
|
|
"num_tokens": 2938754.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.844933986663818,
|
|
"epoch": 1.4164413196322336,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004984404028148585,
|
|
"loss": 5.8044,
|
|
"mean_token_accuracy": 0.16392858922481537,
|
|
"num_tokens": 2949696.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 5.995720243453979,
|
|
"epoch": 1.4218496484586263,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004983895828427062,
|
|
"loss": 5.9963,
|
|
"mean_token_accuracy": 0.1491517722606659,
|
|
"num_tokens": 2961144.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.037636661529541,
|
|
"epoch": 1.427257977285019,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004983379510722935,
|
|
"loss": 5.9821,
|
|
"mean_token_accuracy": 0.15617446154356002,
|
|
"num_tokens": 2972368.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 5.974224662780761,
|
|
"epoch": 1.4326663061114115,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004982855076912506,
|
|
"loss": 5.9083,
|
|
"mean_token_accuracy": 0.15516280233860016,
|
|
"num_tokens": 2984853.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.030554676055909,
|
|
"epoch": 1.4380746349378042,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004982322528901568,
|
|
"loss": 5.9787,
|
|
"mean_token_accuracy": 0.1576838731765747,
|
|
"num_tokens": 2995095.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 5.933174562454224,
|
|
"epoch": 1.4434829637641968,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00049817818686254,
|
|
"loss": 6.0144,
|
|
"mean_token_accuracy": 0.15284789279103278,
|
|
"num_tokens": 3005926.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.117506790161133,
|
|
"epoch": 1.4488912925905895,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004981233098048765,
|
|
"loss": 5.9882,
|
|
"mean_token_accuracy": 0.15670939832925795,
|
|
"num_tokens": 3017647.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 5.873747301101685,
|
|
"epoch": 1.454299621416982,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004980676219165894,
|
|
"loss": 5.8985,
|
|
"mean_token_accuracy": 0.16080968528985978,
|
|
"num_tokens": 3028904.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.097314739227295,
|
|
"epoch": 1.4597079502433747,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000498011123400049,
|
|
"loss": 5.9344,
|
|
"mean_token_accuracy": 0.16150693595409393,
|
|
"num_tokens": 3039347.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 5.886366558074951,
|
|
"epoch": 1.4651162790697674,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004979538144605707,
|
|
"loss": 5.916,
|
|
"mean_token_accuracy": 0.15625391006469727,
|
|
"num_tokens": 3049951.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.073115348815918,
|
|
"epoch": 1.47052460789616,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004978956953064154,
|
|
"loss": 5.9715,
|
|
"mean_token_accuracy": 0.1570509046316147,
|
|
"num_tokens": 3061894.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 5.950630235671997,
|
|
"epoch": 1.4759329367225527,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004978367661487883,
|
|
"loss": 5.9283,
|
|
"mean_token_accuracy": 0.16050151437520982,
|
|
"num_tokens": 3072393.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 5.929414939880371,
|
|
"epoch": 1.4813412655489453,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004977770272018379,
|
|
"loss": 5.9653,
|
|
"mean_token_accuracy": 0.15721761882305146,
|
|
"num_tokens": 3083541.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 5.9391296863555905,
|
|
"epoch": 1.486749594375338,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000497716478682656,
|
|
"loss": 5.7959,
|
|
"mean_token_accuracy": 0.15988317131996155,
|
|
"num_tokens": 3093851.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 5.851752376556396,
|
|
"epoch": 1.4921579232017308,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004976551208112759,
|
|
"loss": 5.8874,
|
|
"mean_token_accuracy": 0.15590570718050004,
|
|
"num_tokens": 3105703.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.07996826171875,
|
|
"epoch": 1.4975662520281232,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004975929538106721,
|
|
"loss": 6.003,
|
|
"mean_token_accuracy": 0.1546159043908119,
|
|
"num_tokens": 3117063.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 5.886177349090576,
|
|
"epoch": 1.502974580854516,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004975299779067598,
|
|
"loss": 5.9831,
|
|
"mean_token_accuracy": 0.15303662568330764,
|
|
"num_tokens": 3129020.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.048104476928711,
|
|
"epoch": 1.5083829096809085,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004974661933283936,
|
|
"loss": 5.9632,
|
|
"mean_token_accuracy": 0.15594938546419143,
|
|
"num_tokens": 3140627.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.00671420097351,
|
|
"epoch": 1.5137912385073014,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004974016003073667,
|
|
"loss": 5.9794,
|
|
"mean_token_accuracy": 0.15382874608039857,
|
|
"num_tokens": 3152294.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 5.957139205932617,
|
|
"epoch": 1.5191995673336938,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004973361990784104,
|
|
"loss": 5.9719,
|
|
"mean_token_accuracy": 0.15267469137907028,
|
|
"num_tokens": 3163770.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.0150069236755375,
|
|
"epoch": 1.5246078961600866,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004972699898791927,
|
|
"loss": 5.9371,
|
|
"mean_token_accuracy": 0.15440668016672135,
|
|
"num_tokens": 3175804.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.996513938903808,
|
|
"epoch": 1.530016224986479,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004972029729503182,
|
|
"loss": 5.9361,
|
|
"mean_token_accuracy": 0.15986578166484833,
|
|
"num_tokens": 3187233.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 5.861429357528687,
|
|
"epoch": 1.535424553812872,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004971351485353265,
|
|
"loss": 5.9091,
|
|
"mean_token_accuracy": 0.16204204708337783,
|
|
"num_tokens": 3198884.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.876031398773193,
|
|
"epoch": 1.5408328826392643,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004970665168806917,
|
|
"loss": 5.8826,
|
|
"mean_token_accuracy": 0.15339980274438858,
|
|
"num_tokens": 3209836.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.9335691928863525,
|
|
"epoch": 1.5462412114656572,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004969970782358214,
|
|
"loss": 5.9038,
|
|
"mean_token_accuracy": 0.15562329590320587,
|
|
"num_tokens": 3221295.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 5.865458536148071,
|
|
"epoch": 1.5516495402920496,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004969268328530557,
|
|
"loss": 5.8515,
|
|
"mean_token_accuracy": 0.16353810727596282,
|
|
"num_tokens": 3231832.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 5.955732727050782,
|
|
"epoch": 1.5570578691184425,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004968557809876666,
|
|
"loss": 5.9523,
|
|
"mean_token_accuracy": 0.1560000017285347,
|
|
"num_tokens": 3243483.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.0030198097229,
|
|
"epoch": 1.562466197944835,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004967839228978568,
|
|
"loss": 5.9858,
|
|
"mean_token_accuracy": 0.1519615024328232,
|
|
"num_tokens": 3255785.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 5.915631580352783,
|
|
"epoch": 1.5678745267712277,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004967112588447585,
|
|
"loss": 5.8646,
|
|
"mean_token_accuracy": 0.1562828913331032,
|
|
"num_tokens": 3267035.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.063388061523438,
|
|
"epoch": 1.5732828555976204,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004966377890924332,
|
|
"loss": 5.9613,
|
|
"mean_token_accuracy": 0.15265796780586244,
|
|
"num_tokens": 3277848.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.840817260742187,
|
|
"epoch": 1.578691184424013,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004965635139078704,
|
|
"loss": 5.8708,
|
|
"mean_token_accuracy": 0.15992295145988464,
|
|
"num_tokens": 3289864.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.03183217048645,
|
|
"epoch": 1.5840995132504057,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004964884335609857,
|
|
"loss": 5.9883,
|
|
"mean_token_accuracy": 0.15850048810243605,
|
|
"num_tokens": 3301441.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.915151309967041,
|
|
"epoch": 1.5895078420767983,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004964125483246218,
|
|
"loss": 5.9452,
|
|
"mean_token_accuracy": 0.15408284813165665,
|
|
"num_tokens": 3313216.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.836554718017578,
|
|
"epoch": 1.594916170903191,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004963358584745457,
|
|
"loss": 5.7689,
|
|
"mean_token_accuracy": 0.15981167703866958,
|
|
"num_tokens": 3323760.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 5.844881772994995,
|
|
"epoch": 1.6003244997295836,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004962583642894482,
|
|
"loss": 5.8462,
|
|
"mean_token_accuracy": 0.15868799537420272,
|
|
"num_tokens": 3334785.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 5.939773464202881,
|
|
"epoch": 1.6057328285559762,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004961800660509437,
|
|
"loss": 5.9094,
|
|
"mean_token_accuracy": 0.16331073641777039,
|
|
"num_tokens": 3345741.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.883348655700684,
|
|
"epoch": 1.6111411573823688,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000496100964043568,
|
|
"loss": 5.8629,
|
|
"mean_token_accuracy": 0.16200875788927077,
|
|
"num_tokens": 3356118.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.883448648452759,
|
|
"epoch": 1.6165494862087615,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000496021058554778,
|
|
"loss": 5.7999,
|
|
"mean_token_accuracy": 0.16425618529319763,
|
|
"num_tokens": 3368107.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.871192264556885,
|
|
"epoch": 1.6219578150351541,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004959403498749505,
|
|
"loss": 5.8179,
|
|
"mean_token_accuracy": 0.1575431525707245,
|
|
"num_tokens": 3379367.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.866976690292359,
|
|
"epoch": 1.6273661438615468,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004958588382973809,
|
|
"loss": 5.9413,
|
|
"mean_token_accuracy": 0.1525462254881859,
|
|
"num_tokens": 3390597.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.993343210220337,
|
|
"epoch": 1.6327744726879394,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004957765241182826,
|
|
"loss": 5.87,
|
|
"mean_token_accuracy": 0.15825925767421722,
|
|
"num_tokens": 3402618.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.701891231536865,
|
|
"epoch": 1.638182801514332,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004956934076367854,
|
|
"loss": 5.7581,
|
|
"mean_token_accuracy": 0.16914322674274446,
|
|
"num_tokens": 3413162.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.902717447280883,
|
|
"epoch": 1.6435911303407247,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000495609489154935,
|
|
"loss": 5.8371,
|
|
"mean_token_accuracy": 0.16880301386117935,
|
|
"num_tokens": 3424331.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.894273567199707,
|
|
"epoch": 1.6489994591671173,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004955247689776911,
|
|
"loss": 5.9644,
|
|
"mean_token_accuracy": 0.15126627832651138,
|
|
"num_tokens": 3434739.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.068590450286865,
|
|
"epoch": 1.65440778799351,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000495439247412927,
|
|
"loss": 6.0419,
|
|
"mean_token_accuracy": 0.1500218965113163,
|
|
"num_tokens": 3445984.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.735113430023193,
|
|
"epoch": 1.6598161168199026,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004953529247714285,
|
|
"loss": 5.7673,
|
|
"mean_token_accuracy": 0.16737849414348602,
|
|
"num_tokens": 3456955.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.058305168151856,
|
|
"epoch": 1.6652244456462952,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004952658013668922,
|
|
"loss": 5.9342,
|
|
"mean_token_accuracy": 0.16117010563611983,
|
|
"num_tokens": 3468895.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.771589946746826,
|
|
"epoch": 1.670632774472688,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004951778775159246,
|
|
"loss": 5.8088,
|
|
"mean_token_accuracy": 0.1627257451415062,
|
|
"num_tokens": 3480083.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.984347581863403,
|
|
"epoch": 1.6760411032990805,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004950891535380414,
|
|
"loss": 5.9634,
|
|
"mean_token_accuracy": 0.15447578579187393,
|
|
"num_tokens": 3492242.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.879773426055908,
|
|
"epoch": 1.6814494321254734,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004949996297556656,
|
|
"loss": 5.7968,
|
|
"mean_token_accuracy": 0.15803838819265364,
|
|
"num_tokens": 3501926.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.802989101409912,
|
|
"epoch": 1.6868577609518658,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004949093064941268,
|
|
"loss": 5.8625,
|
|
"mean_token_accuracy": 0.1666681319475174,
|
|
"num_tokens": 3511421.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.89474196434021,
|
|
"epoch": 1.6922660897782587,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004948181840816601,
|
|
"loss": 5.7942,
|
|
"mean_token_accuracy": 0.16357696503400804,
|
|
"num_tokens": 3522033.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.837677526473999,
|
|
"epoch": 1.697674418604651,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004947262628494045,
|
|
"loss": 5.8414,
|
|
"mean_token_accuracy": 0.1625886917114258,
|
|
"num_tokens": 3533232.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.908598518371582,
|
|
"epoch": 1.703082747431044,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004946335431314018,
|
|
"loss": 5.8576,
|
|
"mean_token_accuracy": 0.1602688431739807,
|
|
"num_tokens": 3544904.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.903112459182739,
|
|
"epoch": 1.7084910762574363,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004945400252645959,
|
|
"loss": 5.9301,
|
|
"mean_token_accuracy": 0.15881284475326538,
|
|
"num_tokens": 3556545.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.889549732208252,
|
|
"epoch": 1.7138994050838292,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004944457095888308,
|
|
"loss": 5.8267,
|
|
"mean_token_accuracy": 0.16299790740013123,
|
|
"num_tokens": 3567361.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.898175001144409,
|
|
"epoch": 1.7193077339102216,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00049435059644685,
|
|
"loss": 5.8407,
|
|
"mean_token_accuracy": 0.16047086715698242,
|
|
"num_tokens": 3578943.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.852223968505859,
|
|
"epoch": 1.7247160627366145,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004942546861842947,
|
|
"loss": 5.8323,
|
|
"mean_token_accuracy": 0.16009423434734343,
|
|
"num_tokens": 3589868.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.86200065612793,
|
|
"epoch": 1.730124391563007,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004941579791497032,
|
|
"loss": 5.8328,
|
|
"mean_token_accuracy": 0.16346521228551864,
|
|
"num_tokens": 3601245.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.853341579437256,
|
|
"epoch": 1.7355327203893998,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004940604756945091,
|
|
"loss": 5.8758,
|
|
"mean_token_accuracy": 0.1621338576078415,
|
|
"num_tokens": 3612168.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.835409259796142,
|
|
"epoch": 1.7409410492157922,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004939621761730399,
|
|
"loss": 5.7899,
|
|
"mean_token_accuracy": 0.16544863879680632,
|
|
"num_tokens": 3624699.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.912599277496338,
|
|
"epoch": 1.746349378042185,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004938630809425166,
|
|
"loss": 5.9346,
|
|
"mean_token_accuracy": 0.1649474322795868,
|
|
"num_tokens": 3635554.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.817644882202148,
|
|
"epoch": 1.7517577068685775,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004937631903630513,
|
|
"loss": 5.8358,
|
|
"mean_token_accuracy": 0.16459024250507354,
|
|
"num_tokens": 3646776.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.833710479736328,
|
|
"epoch": 1.7571660356949703,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004936625047976468,
|
|
"loss": 5.7688,
|
|
"mean_token_accuracy": 0.16519935131073,
|
|
"num_tokens": 3657976.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.925186967849731,
|
|
"epoch": 1.7625743645213627,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004935610246121944,
|
|
"loss": 5.9077,
|
|
"mean_token_accuracy": 0.15773554295301437,
|
|
"num_tokens": 3668242.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.945837879180909,
|
|
"epoch": 1.7679826933477556,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004934587501754736,
|
|
"loss": 5.9204,
|
|
"mean_token_accuracy": 0.16058315932750702,
|
|
"num_tokens": 3679200.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.843207263946534,
|
|
"epoch": 1.7733910221741482,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004933556818591498,
|
|
"loss": 5.8215,
|
|
"mean_token_accuracy": 0.1642814427614212,
|
|
"num_tokens": 3690227.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.9487780094146725,
|
|
"epoch": 1.7787993510005409,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004932518200377736,
|
|
"loss": 5.8698,
|
|
"mean_token_accuracy": 0.16072406321763993,
|
|
"num_tokens": 3701821.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.767692804336548,
|
|
"epoch": 1.7842076798269335,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004931471650887789,
|
|
"loss": 5.7904,
|
|
"mean_token_accuracy": 0.15893402695655823,
|
|
"num_tokens": 3712420.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.939088487625122,
|
|
"epoch": 1.7896160086533262,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004930417173924822,
|
|
"loss": 5.8449,
|
|
"mean_token_accuracy": 0.16310243904590607,
|
|
"num_tokens": 3722211.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.806006813049317,
|
|
"epoch": 1.7950243374797188,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004929354773320806,
|
|
"loss": 5.8313,
|
|
"mean_token_accuracy": 0.16155394315719604,
|
|
"num_tokens": 3732822.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.881654262542725,
|
|
"epoch": 1.8004326663061114,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004928284452936506,
|
|
"loss": 5.8926,
|
|
"mean_token_accuracy": 0.16218796074390412,
|
|
"num_tokens": 3744262.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.851164674758911,
|
|
"epoch": 1.805840995132504,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004927206216661473,
|
|
"loss": 5.7785,
|
|
"mean_token_accuracy": 0.1639172226190567,
|
|
"num_tokens": 3755713.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.768859148025513,
|
|
"epoch": 1.8112493239588967,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004926120068414015,
|
|
"loss": 5.8002,
|
|
"mean_token_accuracy": 0.1638001874089241,
|
|
"num_tokens": 3767179.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.842668867111206,
|
|
"epoch": 1.8166576527852893,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004925026012141201,
|
|
"loss": 5.8107,
|
|
"mean_token_accuracy": 0.1638934686779976,
|
|
"num_tokens": 3777714.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.901233863830567,
|
|
"epoch": 1.822065981611682,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004923924051818832,
|
|
"loss": 5.7871,
|
|
"mean_token_accuracy": 0.1725088521838188,
|
|
"num_tokens": 3788092.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.82500729560852,
|
|
"epoch": 1.8274743104380746,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004922814191451435,
|
|
"loss": 5.7713,
|
|
"mean_token_accuracy": 0.1644965872168541,
|
|
"num_tokens": 3798080.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.848507499694824,
|
|
"epoch": 1.8328826392644673,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004921696435072245,
|
|
"loss": 5.8193,
|
|
"mean_token_accuracy": 0.16445939093828202,
|
|
"num_tokens": 3809547.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.825184059143067,
|
|
"epoch": 1.83829096809086,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004920570786743191,
|
|
"loss": 5.8064,
|
|
"mean_token_accuracy": 0.1676018714904785,
|
|
"num_tokens": 3820681.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.825942611694336,
|
|
"epoch": 1.8436992969172525,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004919437250554882,
|
|
"loss": 5.7937,
|
|
"mean_token_accuracy": 0.165639466047287,
|
|
"num_tokens": 3831598.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.861527681350708,
|
|
"epoch": 1.8491076257436452,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004918295830626592,
|
|
"loss": 5.8498,
|
|
"mean_token_accuracy": 0.16639893054962157,
|
|
"num_tokens": 3843437.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.763908529281617,
|
|
"epoch": 1.8545159545700378,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004917146531106242,
|
|
"loss": 5.7208,
|
|
"mean_token_accuracy": 0.1637628436088562,
|
|
"num_tokens": 3855387.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.838849639892578,
|
|
"epoch": 1.8599242833964305,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004915989356170389,
|
|
"loss": 5.8857,
|
|
"mean_token_accuracy": 0.15771711021661758,
|
|
"num_tokens": 3866999.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.858967590332031,
|
|
"epoch": 1.865332612222823,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004914824310024213,
|
|
"loss": 5.8114,
|
|
"mean_token_accuracy": 0.16324352771043776,
|
|
"num_tokens": 3877721.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.883812522888183,
|
|
"epoch": 1.870740941049216,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004913651396901489,
|
|
"loss": 5.8842,
|
|
"mean_token_accuracy": 0.1559979274868965,
|
|
"num_tokens": 3888831.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.822631740570069,
|
|
"epoch": 1.8761492698756084,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000491247062106459,
|
|
"loss": 5.8255,
|
|
"mean_token_accuracy": 0.16635372191667558,
|
|
"num_tokens": 3900389.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.813898754119873,
|
|
"epoch": 1.8815575987020012,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004911281986804455,
|
|
"loss": 5.7918,
|
|
"mean_token_accuracy": 0.16552133858203888,
|
|
"num_tokens": 3911641.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.699506855010986,
|
|
"epoch": 1.8869659275283936,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004910085498440587,
|
|
"loss": 5.6765,
|
|
"mean_token_accuracy": 0.1682187110185623,
|
|
"num_tokens": 3922877.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.864941835403442,
|
|
"epoch": 1.8923742563547865,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004908881160321025,
|
|
"loss": 5.746,
|
|
"mean_token_accuracy": 0.16923027187585832,
|
|
"num_tokens": 3933573.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.770928573608399,
|
|
"epoch": 1.897782585181179,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004907668976822338,
|
|
"loss": 5.711,
|
|
"mean_token_accuracy": 0.17468960136175155,
|
|
"num_tokens": 3944339.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.912581491470337,
|
|
"epoch": 1.9031909140075718,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004906448952349603,
|
|
"loss": 5.802,
|
|
"mean_token_accuracy": 0.17707422375679016,
|
|
"num_tokens": 3955459.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.685616207122803,
|
|
"epoch": 1.9085992428339642,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004905221091336392,
|
|
"loss": 5.7126,
|
|
"mean_token_accuracy": 0.16650191992521285,
|
|
"num_tokens": 3966782.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.814038848876953,
|
|
"epoch": 1.914007571660357,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004903985398244753,
|
|
"loss": 5.8455,
|
|
"mean_token_accuracy": 0.16468454003334046,
|
|
"num_tokens": 3978794.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.943782377243042,
|
|
"epoch": 1.9194159004867495,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004902741877565201,
|
|
"loss": 5.8506,
|
|
"mean_token_accuracy": 0.16398394256830215,
|
|
"num_tokens": 3990838.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.823399209976197,
|
|
"epoch": 1.9248242293131423,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000490149053381669,
|
|
"loss": 5.8649,
|
|
"mean_token_accuracy": 0.16404231041669845,
|
|
"num_tokens": 4000630.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.888064622879028,
|
|
"epoch": 1.9302325581395348,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004900231371546608,
|
|
"loss": 5.7502,
|
|
"mean_token_accuracy": 0.1688971534371376,
|
|
"num_tokens": 4011433.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.830664825439453,
|
|
"epoch": 1.9356408869659276,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004898964395330753,
|
|
"loss": 5.9134,
|
|
"mean_token_accuracy": 0.16155715361237527,
|
|
"num_tokens": 4023671.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.789358425140381,
|
|
"epoch": 1.94104921579232,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004897689609773318,
|
|
"loss": 5.6873,
|
|
"mean_token_accuracy": 0.17001506835222244,
|
|
"num_tokens": 4034686.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.860217046737671,
|
|
"epoch": 1.946457544618713,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004896407019506879,
|
|
"loss": 5.7935,
|
|
"mean_token_accuracy": 0.1718866765499115,
|
|
"num_tokens": 4046387.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.728650760650635,
|
|
"epoch": 1.9518658734451053,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004895116629192371,
|
|
"loss": 5.7218,
|
|
"mean_token_accuracy": 0.16833948940038682,
|
|
"num_tokens": 4056103.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.776650047302246,
|
|
"epoch": 1.9572742022714982,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004893818443519074,
|
|
"loss": 5.6792,
|
|
"mean_token_accuracy": 0.17679304182529448,
|
|
"num_tokens": 4066940.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.719994163513183,
|
|
"epoch": 1.9626825310978906,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048925124672046,
|
|
"loss": 5.8114,
|
|
"mean_token_accuracy": 0.17301934510469436,
|
|
"num_tokens": 4077232.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.849878931045533,
|
|
"epoch": 1.9680908599242835,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004891198704994868,
|
|
"loss": 5.8796,
|
|
"mean_token_accuracy": 0.16075178533792495,
|
|
"num_tokens": 4089322.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.805488872528076,
|
|
"epoch": 1.973499188750676,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004889877161664096,
|
|
"loss": 5.7189,
|
|
"mean_token_accuracy": 0.17121635377407074,
|
|
"num_tokens": 4099737.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.747078943252563,
|
|
"epoch": 1.9789075175770687,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004888547842014771,
|
|
"loss": 5.7266,
|
|
"mean_token_accuracy": 0.1624513253569603,
|
|
"num_tokens": 4110793.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.807029581069946,
|
|
"epoch": 1.9843158464034614,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004887210750877648,
|
|
"loss": 5.7593,
|
|
"mean_token_accuracy": 0.1650367185473442,
|
|
"num_tokens": 4121481.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.760784578323364,
|
|
"epoch": 1.989724175229854,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004885865893111716,
|
|
"loss": 5.7223,
|
|
"mean_token_accuracy": 0.1747642859816551,
|
|
"num_tokens": 4132701.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.797788667678833,
|
|
"epoch": 1.9951325040562466,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004884513273604195,
|
|
"loss": 5.7938,
|
|
"mean_token_accuracy": 0.1691266641020775,
|
|
"num_tokens": 4143764.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.849646303388807,
|
|
"epoch": 2.0,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 0.0004883152897270502,
|
|
"loss": 5.7515,
|
|
"mean_token_accuracy": 0.16810581419203016,
|
|
"num_tokens": 4153778.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.790167665481567,
|
|
"epoch": 2.005408328826393,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004881784769054253,
|
|
"loss": 5.4643,
|
|
"mean_token_accuracy": 0.17961378693580626,
|
|
"num_tokens": 4165099.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.802452802658081,
|
|
"epoch": 2.0108166576527853,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004880408893927226,
|
|
"loss": 5.5528,
|
|
"mean_token_accuracy": 0.17948687970638275,
|
|
"num_tokens": 4178051.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.761927843093872,
|
|
"epoch": 2.016224986479178,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004879025276889355,
|
|
"loss": 5.4548,
|
|
"mean_token_accuracy": 0.1810762792825699,
|
|
"num_tokens": 4189987.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.764883708953858,
|
|
"epoch": 2.0216333153055706,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048776339229687077,
|
|
"loss": 5.4913,
|
|
"mean_token_accuracy": 0.1828714966773987,
|
|
"num_tokens": 4200999.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.655160522460937,
|
|
"epoch": 2.0270416441319634,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048762348372214683,
|
|
"loss": 5.4144,
|
|
"mean_token_accuracy": 0.1770107254385948,
|
|
"num_tokens": 4212122.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.761785984039307,
|
|
"epoch": 2.032449972958356,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048748280247319165,
|
|
"loss": 5.4894,
|
|
"mean_token_accuracy": 0.17939014881849288,
|
|
"num_tokens": 4222580.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.697722339630127,
|
|
"epoch": 2.0378583017847487,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004873413490612412,
|
|
"loss": 5.3501,
|
|
"mean_token_accuracy": 0.17957222908735276,
|
|
"num_tokens": 4233756.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.667776870727539,
|
|
"epoch": 2.043266630611141,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048719912400033757,
|
|
"loss": 5.4732,
|
|
"mean_token_accuracy": 0.17797181904315948,
|
|
"num_tokens": 4245625.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.830563974380493,
|
|
"epoch": 2.048674959437534,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000487056127807327,
|
|
"loss": 5.6831,
|
|
"mean_token_accuracy": 0.17253895699977875,
|
|
"num_tokens": 4257788.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.6603833675384525,
|
|
"epoch": 2.0540832882639264,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048691236100185795,
|
|
"loss": 5.4369,
|
|
"mean_token_accuracy": 0.1824871927499771,
|
|
"num_tokens": 4269641.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.737438488006592,
|
|
"epoch": 2.0594916170903192,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004867678241063792,
|
|
"loss": 5.4908,
|
|
"mean_token_accuracy": 0.1820552185177803,
|
|
"num_tokens": 4280952.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.721802616119385,
|
|
"epoch": 2.0648999459167117,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048662251764613833,
|
|
"loss": 5.4454,
|
|
"mean_token_accuracy": 0.18466749638319016,
|
|
"num_tokens": 4294371.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.569987487792969,
|
|
"epoch": 2.0703082747431045,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004864764421491793,
|
|
"loss": 5.4462,
|
|
"mean_token_accuracy": 0.17653877884149552,
|
|
"num_tokens": 4307112.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.728185510635376,
|
|
"epoch": 2.075716603569497,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004863295981463408,
|
|
"loss": 5.4179,
|
|
"mean_token_accuracy": 0.1802745908498764,
|
|
"num_tokens": 4317199.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.5756903171539305,
|
|
"epoch": 2.08112493239589,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004861819861712543,
|
|
"loss": 5.4354,
|
|
"mean_token_accuracy": 0.1826627105474472,
|
|
"num_tokens": 4328546.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.781250143051148,
|
|
"epoch": 2.086533261222282,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048603360676034203,
|
|
"loss": 5.4897,
|
|
"mean_token_accuracy": 0.18274689018726348,
|
|
"num_tokens": 4338795.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.597852611541748,
|
|
"epoch": 2.091941590048675,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048588446045281505,
|
|
"loss": 5.4664,
|
|
"mean_token_accuracy": 0.18191643804311752,
|
|
"num_tokens": 4350373.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.669932508468628,
|
|
"epoch": 2.0973499188750675,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004857345477906714,
|
|
"loss": 5.4511,
|
|
"mean_token_accuracy": 0.1787646234035492,
|
|
"num_tokens": 4362801.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.763370227813721,
|
|
"epoch": 2.1027582477014604,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048558386931869417,
|
|
"loss": 5.5536,
|
|
"mean_token_accuracy": 0.1799535408616066,
|
|
"num_tokens": 4375782.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.629010105133057,
|
|
"epoch": 2.1081665765278528,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004854324255844492,
|
|
"loss": 5.4475,
|
|
"mean_token_accuracy": 0.17899309247732162,
|
|
"num_tokens": 4385580.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.688996505737305,
|
|
"epoch": 2.1135749053542456,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048528021713828336,
|
|
"loss": 5.4603,
|
|
"mean_token_accuracy": 0.17725565731525422,
|
|
"num_tokens": 4397434.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.705441951751709,
|
|
"epoch": 2.118983234180638,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004851272445333225,
|
|
"loss": 5.5093,
|
|
"mean_token_accuracy": 0.1757824644446373,
|
|
"num_tokens": 4407888.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.601096248626709,
|
|
"epoch": 2.124391563007031,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048497350832546937,
|
|
"loss": 5.4204,
|
|
"mean_token_accuracy": 0.18114884942770004,
|
|
"num_tokens": 4420010.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.678840303421021,
|
|
"epoch": 2.1297998918334233,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004848190090734018,
|
|
"loss": 5.4594,
|
|
"mean_token_accuracy": 0.17557892948389053,
|
|
"num_tokens": 4430730.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.618314266204834,
|
|
"epoch": 2.135208220659816,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048466374733857035,
|
|
"loss": 5.4402,
|
|
"mean_token_accuracy": 0.18506533801555633,
|
|
"num_tokens": 4442296.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.681797218322754,
|
|
"epoch": 2.1406165494862086,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004845077236851966,
|
|
"loss": 5.4774,
|
|
"mean_token_accuracy": 0.17849523872137069,
|
|
"num_tokens": 4453996.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.6223681449890135,
|
|
"epoch": 2.1460248783126015,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048435093868027076,
|
|
"loss": 5.4834,
|
|
"mean_token_accuracy": 0.1794159531593323,
|
|
"num_tokens": 4465779.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.560337114334106,
|
|
"epoch": 2.151433207138994,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048419339289354995,
|
|
"loss": 5.4092,
|
|
"mean_token_accuracy": 0.18837132751941682,
|
|
"num_tokens": 4477113.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.648620700836181,
|
|
"epoch": 2.1568415359653867,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000484035086897556,
|
|
"loss": 5.3689,
|
|
"mean_token_accuracy": 0.18140988796949387,
|
|
"num_tokens": 4487920.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.6848094940185545,
|
|
"epoch": 2.162249864791779,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004838760212675732,
|
|
"loss": 5.5565,
|
|
"mean_token_accuracy": 0.17392570972442628,
|
|
"num_tokens": 4499049.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.162249864791779,
|
|
"eval_entropy": 5.572800342603164,
|
|
"eval_loss": 6.065261363983154,
|
|
"eval_mean_token_accuracy": 0.1627881208096038,
|
|
"eval_num_tokens": 4499049.0,
|
|
"eval_runtime": 2.7199,
|
|
"eval_samples_per_second": 1292.333,
|
|
"eval_steps_per_second": 161.771,
|
|
"step": 2000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9240,
|
|
"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": 7052141297664000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|