8079 lines
219 KiB
JSON
8079 lines
219 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 4.324499729583558,
|
|
"eval_steps": 1000,
|
|
"global_step": 4000,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 5.645669031143188,
|
|
"epoch": 2.167658193618172,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004837161965816464,
|
|
"loss": 5.3685,
|
|
"mean_token_accuracy": 0.18813167363405228,
|
|
"num_tokens": 4508944.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.646870470046997,
|
|
"epoch": 2.1730665224445644,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048355561342057886,
|
|
"loss": 5.4151,
|
|
"mean_token_accuracy": 0.17982712686061858,
|
|
"num_tokens": 4518845.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.509386396408081,
|
|
"epoch": 2.1784748512709573,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048339427236793024,
|
|
"loss": 5.4707,
|
|
"mean_token_accuracy": 0.17715146243572236,
|
|
"num_tokens": 4530416.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.696588706970215,
|
|
"epoch": 2.1838831800973497,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004832321740100143,
|
|
"loss": 5.4502,
|
|
"mean_token_accuracy": 0.18018950819969176,
|
|
"num_tokens": 4542630.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.536802291870117,
|
|
"epoch": 2.1892915089237426,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048306931893589666,
|
|
"loss": 5.4528,
|
|
"mean_token_accuracy": 0.17790227085351945,
|
|
"num_tokens": 4553270.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.531373500823975,
|
|
"epoch": 2.194699837750135,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048290570773739327,
|
|
"loss": 5.3122,
|
|
"mean_token_accuracy": 0.1856985941529274,
|
|
"num_tokens": 4564647.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.660735940933227,
|
|
"epoch": 2.200108166576528,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048274134100906747,
|
|
"loss": 5.4901,
|
|
"mean_token_accuracy": 0.17462403774261476,
|
|
"num_tokens": 4575507.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.644405460357666,
|
|
"epoch": 2.2055164954029207,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00048257621934822835,
|
|
"loss": 5.477,
|
|
"mean_token_accuracy": 0.19022988080978392,
|
|
"num_tokens": 4586761.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.539322423934936,
|
|
"epoch": 2.210924824229313,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004824103433549285,
|
|
"loss": 5.4003,
|
|
"mean_token_accuracy": 0.18904444575309753,
|
|
"num_tokens": 4597689.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.687419319152832,
|
|
"epoch": 2.2163331530557056,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004822437136319616,
|
|
"loss": 5.4493,
|
|
"mean_token_accuracy": 0.17900420874357223,
|
|
"num_tokens": 4608886.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.625580310821533,
|
|
"epoch": 2.2217414818820984,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004820763307848604,
|
|
"loss": 5.5054,
|
|
"mean_token_accuracy": 0.17139943391084672,
|
|
"num_tokens": 4619989.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.575812578201294,
|
|
"epoch": 2.2271498107084913,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048190819542189456,
|
|
"loss": 5.373,
|
|
"mean_token_accuracy": 0.18371485024690629,
|
|
"num_tokens": 4630218.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.58152322769165,
|
|
"epoch": 2.2325581395348837,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048173930815406847,
|
|
"loss": 5.468,
|
|
"mean_token_accuracy": 0.17635657489299775,
|
|
"num_tokens": 4642024.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.641052532196045,
|
|
"epoch": 2.2379664683612766,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004815696695951187,
|
|
"loss": 5.487,
|
|
"mean_token_accuracy": 0.18174674063920976,
|
|
"num_tokens": 4653370.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.632121849060058,
|
|
"epoch": 2.243374797187669,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004813992803615123,
|
|
"loss": 5.4673,
|
|
"mean_token_accuracy": 0.18117746263742446,
|
|
"num_tokens": 4664522.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.648852968215943,
|
|
"epoch": 2.248783126014062,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048122814107244414,
|
|
"loss": 5.5202,
|
|
"mean_token_accuracy": 0.180680613219738,
|
|
"num_tokens": 4676542.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.528378963470459,
|
|
"epoch": 2.2541914548404542,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048105625234983466,
|
|
"loss": 5.4388,
|
|
"mean_token_accuracy": 0.18309595733880996,
|
|
"num_tokens": 4687810.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.7000439167022705,
|
|
"epoch": 2.259599783666847,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048088361481832807,
|
|
"loss": 5.4124,
|
|
"mean_token_accuracy": 0.18323352932929993,
|
|
"num_tokens": 4697661.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.530203628540039,
|
|
"epoch": 2.2650081124932395,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048071022910528934,
|
|
"loss": 5.3657,
|
|
"mean_token_accuracy": 0.18107171952724457,
|
|
"num_tokens": 4708585.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.667945098876953,
|
|
"epoch": 2.2704164413196324,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048053609584080274,
|
|
"loss": 5.5189,
|
|
"mean_token_accuracy": 0.18629098534584046,
|
|
"num_tokens": 4719848.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.633552980422974,
|
|
"epoch": 2.275824770146025,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048036121565766876,
|
|
"loss": 5.5112,
|
|
"mean_token_accuracy": 0.17468099892139435,
|
|
"num_tokens": 4731474.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.591448783874512,
|
|
"epoch": 2.2812330989724177,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004801855891914026,
|
|
"loss": 5.4785,
|
|
"mean_token_accuracy": 0.18278463780879975,
|
|
"num_tokens": 4743081.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.639946603775025,
|
|
"epoch": 2.28664142779881,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048000921708023117,
|
|
"loss": 5.4567,
|
|
"mean_token_accuracy": 0.17776891738176345,
|
|
"num_tokens": 4754829.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.5485023021697994,
|
|
"epoch": 2.292049756625203,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004798320999650911,
|
|
"loss": 5.4287,
|
|
"mean_token_accuracy": 0.18394529372453688,
|
|
"num_tokens": 4765522.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.594610071182251,
|
|
"epoch": 2.2974580854515954,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047965423848962636,
|
|
"loss": 5.4097,
|
|
"mean_token_accuracy": 0.18680063188076018,
|
|
"num_tokens": 4776246.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.579386854171753,
|
|
"epoch": 2.302866414277988,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000479475633300186,
|
|
"loss": 5.4772,
|
|
"mean_token_accuracy": 0.1775058388710022,
|
|
"num_tokens": 4787595.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.695283269882202,
|
|
"epoch": 2.3082747431043806,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047929628504582173,
|
|
"loss": 5.5139,
|
|
"mean_token_accuracy": 0.17357533425092697,
|
|
"num_tokens": 4798877.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.607059669494629,
|
|
"epoch": 2.3136830719307735,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004791161943782856,
|
|
"loss": 5.5202,
|
|
"mean_token_accuracy": 0.18035576790571212,
|
|
"num_tokens": 4809542.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.732069492340088,
|
|
"epoch": 2.319091400757166,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047893536195202726,
|
|
"loss": 5.6316,
|
|
"mean_token_accuracy": 0.17139614075422288,
|
|
"num_tokens": 4820982.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.527982378005982,
|
|
"epoch": 2.3244997295835588,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004787537884241923,
|
|
"loss": 5.4658,
|
|
"mean_token_accuracy": 0.17957597076892853,
|
|
"num_tokens": 4831543.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.835995817184449,
|
|
"epoch": 2.329908058409951,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047857147445461943,
|
|
"loss": 5.526,
|
|
"mean_token_accuracy": 0.17487936913967134,
|
|
"num_tokens": 4843221.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.5141805648803714,
|
|
"epoch": 2.335316387236344,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004783884207058379,
|
|
"loss": 5.3996,
|
|
"mean_token_accuracy": 0.18044232279062272,
|
|
"num_tokens": 4855201.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.649737119674683,
|
|
"epoch": 2.3407247160627365,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047820462784306555,
|
|
"loss": 5.5013,
|
|
"mean_token_accuracy": 0.18132029622793197,
|
|
"num_tokens": 4866387.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.5604620456695555,
|
|
"epoch": 2.3461330448891293,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000478020096534206,
|
|
"loss": 5.3996,
|
|
"mean_token_accuracy": 0.18129459172487258,
|
|
"num_tokens": 4877702.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.584481048583984,
|
|
"epoch": 2.3515413737155217,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047783482744984644,
|
|
"loss": 5.4797,
|
|
"mean_token_accuracy": 0.17923865616321563,
|
|
"num_tokens": 4888652.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.680005836486816,
|
|
"epoch": 2.3569497025419146,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047764882126325527,
|
|
"loss": 5.5868,
|
|
"mean_token_accuracy": 0.17480754256248474,
|
|
"num_tokens": 4900995.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.561703395843506,
|
|
"epoch": 2.362358031368307,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004774620786503793,
|
|
"loss": 5.3536,
|
|
"mean_token_accuracy": 0.18677263110876083,
|
|
"num_tokens": 4912173.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.486861896514893,
|
|
"epoch": 2.3677663601947,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004772746002898417,
|
|
"loss": 5.3903,
|
|
"mean_token_accuracy": 0.1850423276424408,
|
|
"num_tokens": 4923149.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.614479398727417,
|
|
"epoch": 2.3731746890210923,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047708638686293903,
|
|
"loss": 5.3923,
|
|
"mean_token_accuracy": 0.18675517290830612,
|
|
"num_tokens": 4934156.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.642486238479615,
|
|
"epoch": 2.378583017847485,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047689743905363963,
|
|
"loss": 5.5845,
|
|
"mean_token_accuracy": 0.1755848005414009,
|
|
"num_tokens": 4946690.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.66228084564209,
|
|
"epoch": 2.383991346673878,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004767077575485801,
|
|
"loss": 5.4932,
|
|
"mean_token_accuracy": 0.18646242916584016,
|
|
"num_tokens": 4958208.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.511665296554566,
|
|
"epoch": 2.3893996755002704,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004765173430370635,
|
|
"loss": 5.4322,
|
|
"mean_token_accuracy": 0.18511777520179748,
|
|
"num_tokens": 4968815.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.623476982116699,
|
|
"epoch": 2.394808004326663,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004763261962110567,
|
|
"loss": 5.4185,
|
|
"mean_token_accuracy": 0.188280887901783,
|
|
"num_tokens": 4979468.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.59236102104187,
|
|
"epoch": 2.4002163331530557,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004761343177651875,
|
|
"loss": 5.5108,
|
|
"mean_token_accuracy": 0.1757075935602188,
|
|
"num_tokens": 4990841.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.708037710189819,
|
|
"epoch": 2.4056246619794486,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004759417083967428,
|
|
"loss": 5.5361,
|
|
"mean_token_accuracy": 0.17709264159202576,
|
|
"num_tokens": 5001567.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.570716667175293,
|
|
"epoch": 2.411032990805841,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047574836880566544,
|
|
"loss": 5.4172,
|
|
"mean_token_accuracy": 0.18018932193517684,
|
|
"num_tokens": 5013219.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.5442201614379885,
|
|
"epoch": 2.4164413196322334,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004755542996945519,
|
|
"loss": 5.5355,
|
|
"mean_token_accuracy": 0.1823438435792923,
|
|
"num_tokens": 5025771.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.624157476425171,
|
|
"epoch": 2.4218496484586263,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047535950176864976,
|
|
"loss": 5.4996,
|
|
"mean_token_accuracy": 0.17955673038959502,
|
|
"num_tokens": 5038205.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.618866014480591,
|
|
"epoch": 2.427257977285019,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047516397573585525,
|
|
"loss": 5.4594,
|
|
"mean_token_accuracy": 0.18345661759376525,
|
|
"num_tokens": 5048428.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.549601459503174,
|
|
"epoch": 2.4326663061114115,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047496772230671034,
|
|
"loss": 5.4313,
|
|
"mean_token_accuracy": 0.18356556296348572,
|
|
"num_tokens": 5058391.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.527605628967285,
|
|
"epoch": 2.4380746349378044,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004747707421944004,
|
|
"loss": 5.469,
|
|
"mean_token_accuracy": 0.18059248179197313,
|
|
"num_tokens": 5069832.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.64908504486084,
|
|
"epoch": 2.443482963764197,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004745730361147517,
|
|
"loss": 5.446,
|
|
"mean_token_accuracy": 0.1879955932497978,
|
|
"num_tokens": 5080198.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.585839700698853,
|
|
"epoch": 2.4488912925905897,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004743746047862286,
|
|
"loss": 5.5144,
|
|
"mean_token_accuracy": 0.1745617300271988,
|
|
"num_tokens": 5090817.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.555514764785767,
|
|
"epoch": 2.454299621416982,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004741754489299309,
|
|
"loss": 5.404,
|
|
"mean_token_accuracy": 0.18054056018590928,
|
|
"num_tokens": 5102699.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.576645946502685,
|
|
"epoch": 2.459707950243375,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047397556926959155,
|
|
"loss": 5.3867,
|
|
"mean_token_accuracy": 0.1779182270169258,
|
|
"num_tokens": 5113922.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.607376956939698,
|
|
"epoch": 2.4651162790697674,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047377496653157374,
|
|
"loss": 5.5075,
|
|
"mean_token_accuracy": 0.17765306532382966,
|
|
"num_tokens": 5124920.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.563904428482056,
|
|
"epoch": 2.4705246078961602,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047357364144486806,
|
|
"loss": 5.4942,
|
|
"mean_token_accuracy": 0.17023996859788895,
|
|
"num_tokens": 5136304.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.627230882644653,
|
|
"epoch": 2.4759329367225527,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047337159474109057,
|
|
"loss": 5.5659,
|
|
"mean_token_accuracy": 0.1758327677845955,
|
|
"num_tokens": 5148101.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.64640851020813,
|
|
"epoch": 2.4813412655489455,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047316882715447934,
|
|
"loss": 5.4857,
|
|
"mean_token_accuracy": 0.18239477574825286,
|
|
"num_tokens": 5159186.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.519102954864502,
|
|
"epoch": 2.486749594375338,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004729653394218922,
|
|
"loss": 5.4683,
|
|
"mean_token_accuracy": 0.179428368806839,
|
|
"num_tokens": 5170542.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.601918888092041,
|
|
"epoch": 2.492157923201731,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047276113228280427,
|
|
"loss": 5.5517,
|
|
"mean_token_accuracy": 0.17589437514543532,
|
|
"num_tokens": 5182685.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.612433671951294,
|
|
"epoch": 2.497566252028123,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004725562064793045,
|
|
"loss": 5.3858,
|
|
"mean_token_accuracy": 0.18306297361850737,
|
|
"num_tokens": 5192622.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.615820217132568,
|
|
"epoch": 2.502974580854516,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047235056275609397,
|
|
"loss": 5.5587,
|
|
"mean_token_accuracy": 0.17594667077064513,
|
|
"num_tokens": 5205068.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.5200278759002686,
|
|
"epoch": 2.5083829096809085,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004721442018604822,
|
|
"loss": 5.4402,
|
|
"mean_token_accuracy": 0.1808630719780922,
|
|
"num_tokens": 5216049.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.5477588176727295,
|
|
"epoch": 2.5137912385073014,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004719371245423853,
|
|
"loss": 5.4288,
|
|
"mean_token_accuracy": 0.1794813722372055,
|
|
"num_tokens": 5227024.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.65534987449646,
|
|
"epoch": 2.5191995673336938,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047172933155432273,
|
|
"loss": 5.5372,
|
|
"mean_token_accuracy": 0.1781221240758896,
|
|
"num_tokens": 5238862.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.6297181129455565,
|
|
"epoch": 2.5246078961600866,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047152082365141466,
|
|
"loss": 5.5592,
|
|
"mean_token_accuracy": 0.1701161116361618,
|
|
"num_tokens": 5250020.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.633702325820923,
|
|
"epoch": 2.530016224986479,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004713116015913794,
|
|
"loss": 5.4849,
|
|
"mean_token_accuracy": 0.1720586523413658,
|
|
"num_tokens": 5260281.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.664947414398194,
|
|
"epoch": 2.535424553812872,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047110166613453016,
|
|
"loss": 5.6033,
|
|
"mean_token_accuracy": 0.17747218459844588,
|
|
"num_tokens": 5272909.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.504727125167847,
|
|
"epoch": 2.5408328826392643,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047089101804377305,
|
|
"loss": 5.4142,
|
|
"mean_token_accuracy": 0.18442951887845993,
|
|
"num_tokens": 5285629.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.694669103622436,
|
|
"epoch": 2.546241211465657,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004706796580846037,
|
|
"loss": 5.478,
|
|
"mean_token_accuracy": 0.18136337846517564,
|
|
"num_tokens": 5296104.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.7222755432128904,
|
|
"epoch": 2.5516495402920496,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047046758702510465,
|
|
"loss": 5.6081,
|
|
"mean_token_accuracy": 0.1727263256907463,
|
|
"num_tokens": 5307148.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.532242107391357,
|
|
"epoch": 2.5570578691184425,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047025480563594264,
|
|
"loss": 5.4329,
|
|
"mean_token_accuracy": 0.17678077071905135,
|
|
"num_tokens": 5317878.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.643880319595337,
|
|
"epoch": 2.5624661979448353,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047004131469036576,
|
|
"loss": 5.515,
|
|
"mean_token_accuracy": 0.17565158158540725,
|
|
"num_tokens": 5329603.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.530762529373169,
|
|
"epoch": 2.5678745267712277,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046982711496420057,
|
|
"loss": 5.4856,
|
|
"mean_token_accuracy": 0.1789496660232544,
|
|
"num_tokens": 5340611.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.573900604248047,
|
|
"epoch": 2.57328285559762,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046961220723584925,
|
|
"loss": 5.3762,
|
|
"mean_token_accuracy": 0.18679004907608032,
|
|
"num_tokens": 5352017.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.591924476623535,
|
|
"epoch": 2.578691184424013,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00046939659228628725,
|
|
"loss": 5.4575,
|
|
"mean_token_accuracy": 0.1829191952943802,
|
|
"num_tokens": 5362482.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.526250410079956,
|
|
"epoch": 2.584099513250406,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00046918027089905956,
|
|
"loss": 5.4576,
|
|
"mean_token_accuracy": 0.1830848142504692,
|
|
"num_tokens": 5374074.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.538753604888916,
|
|
"epoch": 2.5895078420767983,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046896324386027873,
|
|
"loss": 5.281,
|
|
"mean_token_accuracy": 0.19114795327186584,
|
|
"num_tokens": 5384650.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.453603363037109,
|
|
"epoch": 2.5949161709031907,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046874551195862145,
|
|
"loss": 5.5003,
|
|
"mean_token_accuracy": 0.17872906923294068,
|
|
"num_tokens": 5395184.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.532912969589233,
|
|
"epoch": 2.6003244997295836,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000468527075985326,
|
|
"loss": 5.3402,
|
|
"mean_token_accuracy": 0.18636347055435182,
|
|
"num_tokens": 5405947.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.541653251647949,
|
|
"epoch": 2.6057328285559764,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00046830793673418924,
|
|
"loss": 5.4477,
|
|
"mean_token_accuracy": 0.18255904763936998,
|
|
"num_tokens": 5418237.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.533209133148193,
|
|
"epoch": 2.611141157382369,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046808809500156355,
|
|
"loss": 5.4845,
|
|
"mean_token_accuracy": 0.17416288554668427,
|
|
"num_tokens": 5429336.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.582285451889038,
|
|
"epoch": 2.6165494862087613,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004678675515863545,
|
|
"loss": 5.4552,
|
|
"mean_token_accuracy": 0.17475373595952987,
|
|
"num_tokens": 5440257.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.657260799407959,
|
|
"epoch": 2.621957815035154,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004676463072900173,
|
|
"loss": 5.5813,
|
|
"mean_token_accuracy": 0.1749948099255562,
|
|
"num_tokens": 5453079.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.5308197975158695,
|
|
"epoch": 2.627366143861547,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046742436291655415,
|
|
"loss": 5.4466,
|
|
"mean_token_accuracy": 0.18183275163173676,
|
|
"num_tokens": 5464551.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.616491079330444,
|
|
"epoch": 2.6327744726879394,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046720171927251155,
|
|
"loss": 5.4832,
|
|
"mean_token_accuracy": 0.1835294097661972,
|
|
"num_tokens": 5474797.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.519690656661988,
|
|
"epoch": 2.638182801514332,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004669783771669769,
|
|
"loss": 5.4227,
|
|
"mean_token_accuracy": 0.18592759519815444,
|
|
"num_tokens": 5485760.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.58955945968628,
|
|
"epoch": 2.6435911303407247,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000466754337411576,
|
|
"loss": 5.4083,
|
|
"mean_token_accuracy": 0.18650511652231216,
|
|
"num_tokens": 5496182.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.54943733215332,
|
|
"epoch": 2.6489994591671175,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046652960082046985,
|
|
"loss": 5.5085,
|
|
"mean_token_accuracy": 0.18460334986448287,
|
|
"num_tokens": 5506848.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.6907923221588135,
|
|
"epoch": 2.65440778799351,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046630416821035164,
|
|
"loss": 5.6166,
|
|
"mean_token_accuracy": 0.17665595561265945,
|
|
"num_tokens": 5518430.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.591123390197754,
|
|
"epoch": 2.6598161168199024,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004660780404004441,
|
|
"loss": 5.4815,
|
|
"mean_token_accuracy": 0.17640893310308456,
|
|
"num_tokens": 5529741.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.504209232330322,
|
|
"epoch": 2.6652244456462952,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046585121821249623,
|
|
"loss": 5.4809,
|
|
"mean_token_accuracy": 0.18376706093549727,
|
|
"num_tokens": 5541835.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.590335178375244,
|
|
"epoch": 2.670632774472688,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046562370247078025,
|
|
"loss": 5.3651,
|
|
"mean_token_accuracy": 0.18910403400659562,
|
|
"num_tokens": 5552177.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.574373435974121,
|
|
"epoch": 2.6760411032990805,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000465395494002089,
|
|
"loss": 5.5876,
|
|
"mean_token_accuracy": 0.1673454150557518,
|
|
"num_tokens": 5564440.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.616112184524536,
|
|
"epoch": 2.6814494321254734,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004651665936357325,
|
|
"loss": 5.4021,
|
|
"mean_token_accuracy": 0.18786526918411256,
|
|
"num_tokens": 5574905.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.566537666320801,
|
|
"epoch": 2.686857760951866,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004649370022035354,
|
|
"loss": 5.4745,
|
|
"mean_token_accuracy": 0.1735750526189804,
|
|
"num_tokens": 5585583.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.52632622718811,
|
|
"epoch": 2.6922660897782587,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004647067205398332,
|
|
"loss": 5.4634,
|
|
"mean_token_accuracy": 0.1833536371588707,
|
|
"num_tokens": 5597077.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.549356603622437,
|
|
"epoch": 2.697674418604651,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046447574948147027,
|
|
"loss": 5.311,
|
|
"mean_token_accuracy": 0.18782647252082824,
|
|
"num_tokens": 5607904.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.499818134307861,
|
|
"epoch": 2.703082747431044,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00046424408986779587,
|
|
"loss": 5.494,
|
|
"mean_token_accuracy": 0.18189319968223572,
|
|
"num_tokens": 5619611.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.635709953308106,
|
|
"epoch": 2.7084910762574363,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004640117425406615,
|
|
"loss": 5.5022,
|
|
"mean_token_accuracy": 0.18441449701786042,
|
|
"num_tokens": 5630365.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.55071268081665,
|
|
"epoch": 2.713899405083829,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000463778708344418,
|
|
"loss": 5.4564,
|
|
"mean_token_accuracy": 0.18218460232019423,
|
|
"num_tokens": 5641008.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.577529239654541,
|
|
"epoch": 2.7193077339102216,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046354498812591207,
|
|
"loss": 5.4889,
|
|
"mean_token_accuracy": 0.18289800584316254,
|
|
"num_tokens": 5652633.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.567134189605713,
|
|
"epoch": 2.7247160627366145,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046331058273448364,
|
|
"loss": 5.4936,
|
|
"mean_token_accuracy": 0.16897694021463394,
|
|
"num_tokens": 5664108.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.59939227104187,
|
|
"epoch": 2.730124391563007,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004630754930219624,
|
|
"loss": 5.4528,
|
|
"mean_token_accuracy": 0.18806864619255065,
|
|
"num_tokens": 5675701.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.410060596466065,
|
|
"epoch": 2.7355327203893998,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004628397198426648,
|
|
"loss": 5.3837,
|
|
"mean_token_accuracy": 0.18483911752700805,
|
|
"num_tokens": 5687292.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.594952154159546,
|
|
"epoch": 2.740941049215792,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046260326405339126,
|
|
"loss": 5.437,
|
|
"mean_token_accuracy": 0.18638764172792435,
|
|
"num_tokens": 5698905.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.46495304107666,
|
|
"epoch": 2.746349378042185,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004623661265134226,
|
|
"loss": 5.3309,
|
|
"mean_token_accuracy": 0.1876953586935997,
|
|
"num_tokens": 5709678.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.5733133316040036,
|
|
"epoch": 2.7517577068685775,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004621283080845172,
|
|
"loss": 5.4479,
|
|
"mean_token_accuracy": 0.18793380111455918,
|
|
"num_tokens": 5720460.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.470430517196656,
|
|
"epoch": 2.7571660356949703,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046188980963090783,
|
|
"loss": 5.475,
|
|
"mean_token_accuracy": 0.17650179862976073,
|
|
"num_tokens": 5731701.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.5680547714233395,
|
|
"epoch": 2.7625743645213627,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046165063201929825,
|
|
"loss": 5.4703,
|
|
"mean_token_accuracy": 0.17936647385358812,
|
|
"num_tokens": 5741512.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.550040197372437,
|
|
"epoch": 2.7679826933477556,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00046141077611886074,
|
|
"loss": 5.3958,
|
|
"mean_token_accuracy": 0.18380182534456252,
|
|
"num_tokens": 5751882.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.51102089881897,
|
|
"epoch": 2.773391022174148,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004611702428012319,
|
|
"loss": 5.5032,
|
|
"mean_token_accuracy": 0.1757590189576149,
|
|
"num_tokens": 5761989.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.574033975601196,
|
|
"epoch": 2.778799351000541,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046092903294051057,
|
|
"loss": 5.4067,
|
|
"mean_token_accuracy": 0.17901558578014373,
|
|
"num_tokens": 5772532.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.602425622940063,
|
|
"epoch": 2.7842076798269337,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046068714741325385,
|
|
"loss": 5.5324,
|
|
"mean_token_accuracy": 0.1822599545121193,
|
|
"num_tokens": 5784152.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.582559537887573,
|
|
"epoch": 2.789616008653326,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004604445870984743,
|
|
"loss": 5.4736,
|
|
"mean_token_accuracy": 0.17780030965805055,
|
|
"num_tokens": 5793748.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.463177680969238,
|
|
"epoch": 2.7950243374797186,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004602013528776367,
|
|
"loss": 5.4149,
|
|
"mean_token_accuracy": 0.1806993380188942,
|
|
"num_tokens": 5804371.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.556249761581421,
|
|
"epoch": 2.8004326663061114,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004599574456346548,
|
|
"loss": 5.4325,
|
|
"mean_token_accuracy": 0.18189264684915543,
|
|
"num_tokens": 5815470.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.594823980331421,
|
|
"epoch": 2.8058409951325043,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045971286625588796,
|
|
"loss": 5.5269,
|
|
"mean_token_accuracy": 0.18027465492486955,
|
|
"num_tokens": 5826037.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.509397220611572,
|
|
"epoch": 2.8112493239588967,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004594676156301383,
|
|
"loss": 5.3925,
|
|
"mean_token_accuracy": 0.18461304306983947,
|
|
"num_tokens": 5836468.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.520547580718994,
|
|
"epoch": 2.816657652785289,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045922169464864725,
|
|
"loss": 5.3107,
|
|
"mean_token_accuracy": 0.18723224252462387,
|
|
"num_tokens": 5846371.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.357919645309448,
|
|
"epoch": 2.822065981611682,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000458975104205092,
|
|
"loss": 5.3872,
|
|
"mean_token_accuracy": 0.18675739467144012,
|
|
"num_tokens": 5856890.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.5643798351287845,
|
|
"epoch": 2.827474310438075,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045872784519558304,
|
|
"loss": 5.4465,
|
|
"mean_token_accuracy": 0.18393590301275253,
|
|
"num_tokens": 5868075.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.633454132080078,
|
|
"epoch": 2.8328826392644673,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004584799185186599,
|
|
"loss": 5.5346,
|
|
"mean_token_accuracy": 0.18316996321082116,
|
|
"num_tokens": 5879579.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.498909378051758,
|
|
"epoch": 2.8382909680908597,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045823132507528903,
|
|
"loss": 5.3777,
|
|
"mean_token_accuracy": 0.18091779947280884,
|
|
"num_tokens": 5890648.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.445182609558105,
|
|
"epoch": 2.8436992969172525,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004579820657688593,
|
|
"loss": 5.4234,
|
|
"mean_token_accuracy": 0.181607748568058,
|
|
"num_tokens": 5902914.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.603010797500611,
|
|
"epoch": 2.8491076257436454,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004577321415051798,
|
|
"loss": 5.4527,
|
|
"mean_token_accuracy": 0.17395040690898894,
|
|
"num_tokens": 5915801.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.617675542831421,
|
|
"epoch": 2.854515954570038,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045748155319247564,
|
|
"loss": 5.5247,
|
|
"mean_token_accuracy": 0.17704819589853288,
|
|
"num_tokens": 5928973.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.559351634979248,
|
|
"epoch": 2.8599242833964302,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004572303017413856,
|
|
"loss": 5.5444,
|
|
"mean_token_accuracy": 0.177216075360775,
|
|
"num_tokens": 5939720.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.582284879684448,
|
|
"epoch": 2.865332612222823,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004569783880649578,
|
|
"loss": 5.4857,
|
|
"mean_token_accuracy": 0.17758719176054,
|
|
"num_tokens": 5951956.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.635481309890747,
|
|
"epoch": 2.870740941049216,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004567258130786474,
|
|
"loss": 5.6355,
|
|
"mean_token_accuracy": 0.17123274654150009,
|
|
"num_tokens": 5963735.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.6331993579864506,
|
|
"epoch": 2.8761492698756084,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045647257770031215,
|
|
"loss": 5.3902,
|
|
"mean_token_accuracy": 0.183367021381855,
|
|
"num_tokens": 5974567.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.513382434844971,
|
|
"epoch": 2.8815575987020012,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045621868285021023,
|
|
"loss": 5.5108,
|
|
"mean_token_accuracy": 0.1816274881362915,
|
|
"num_tokens": 5985800.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.478516626358032,
|
|
"epoch": 2.8869659275283936,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045596412945099615,
|
|
"loss": 5.3655,
|
|
"mean_token_accuracy": 0.18536760807037353,
|
|
"num_tokens": 5997388.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.590605878829956,
|
|
"epoch": 2.8923742563547865,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045570891842771727,
|
|
"loss": 5.4698,
|
|
"mean_token_accuracy": 0.18438928276300431,
|
|
"num_tokens": 6008739.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.456298589706421,
|
|
"epoch": 2.897782585181179,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045545305070781135,
|
|
"loss": 5.358,
|
|
"mean_token_accuracy": 0.1847207322716713,
|
|
"num_tokens": 6019490.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.501442575454712,
|
|
"epoch": 2.903190914007572,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045519652722110206,
|
|
"loss": 5.4838,
|
|
"mean_token_accuracy": 0.1852235645055771,
|
|
"num_tokens": 6030712.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.581159973144532,
|
|
"epoch": 2.908599242833964,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045493934889979655,
|
|
"loss": 5.3725,
|
|
"mean_token_accuracy": 0.18857915252447127,
|
|
"num_tokens": 6042161.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.462815237045288,
|
|
"epoch": 2.914007571660357,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045468151667848116,
|
|
"loss": 5.3945,
|
|
"mean_token_accuracy": 0.1814683720469475,
|
|
"num_tokens": 6054107.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.520541667938232,
|
|
"epoch": 2.9194159004867495,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000454423031494119,
|
|
"loss": 5.4664,
|
|
"mean_token_accuracy": 0.18704262524843215,
|
|
"num_tokens": 6065232.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.535909700393677,
|
|
"epoch": 2.9248242293131423,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004541638942860458,
|
|
"loss": 5.4452,
|
|
"mean_token_accuracy": 0.17806587219238282,
|
|
"num_tokens": 6076294.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.477101612091064,
|
|
"epoch": 2.9302325581395348,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004539041059959667,
|
|
"loss": 5.3473,
|
|
"mean_token_accuracy": 0.19883838742971421,
|
|
"num_tokens": 6087700.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.5099649906158445,
|
|
"epoch": 2.9356408869659276,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045364366756795295,
|
|
"loss": 5.4288,
|
|
"mean_token_accuracy": 0.18620026260614395,
|
|
"num_tokens": 6098561.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.6038124561309814,
|
|
"epoch": 2.94104921579232,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045338257994843845,
|
|
"loss": 5.4688,
|
|
"mean_token_accuracy": 0.18296482115983964,
|
|
"num_tokens": 6109729.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.500819826126099,
|
|
"epoch": 2.946457544618713,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004531208440862162,
|
|
"loss": 5.4351,
|
|
"mean_token_accuracy": 0.18409371972084046,
|
|
"num_tokens": 6120468.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.487791061401367,
|
|
"epoch": 2.9518658734451053,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00045285846093243485,
|
|
"loss": 5.3663,
|
|
"mean_token_accuracy": 0.18640440404415132,
|
|
"num_tokens": 6132457.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.458179092407226,
|
|
"epoch": 2.957274202271498,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045259543144059557,
|
|
"loss": 5.4263,
|
|
"mean_token_accuracy": 0.18648390471935272,
|
|
"num_tokens": 6142289.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.553471326828003,
|
|
"epoch": 2.9626825310978906,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004523317565665479,
|
|
"loss": 5.4241,
|
|
"mean_token_accuracy": 0.18614060878753663,
|
|
"num_tokens": 6153786.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.561613750457764,
|
|
"epoch": 2.9680908599242835,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045206743726848716,
|
|
"loss": 5.4214,
|
|
"mean_token_accuracy": 0.1874699130654335,
|
|
"num_tokens": 6163988.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.440897750854492,
|
|
"epoch": 2.973499188750676,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004518024745069502,
|
|
"loss": 5.513,
|
|
"mean_token_accuracy": 0.1822364389896393,
|
|
"num_tokens": 6175859.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.812635660171509,
|
|
"epoch": 2.9789075175770687,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004515368692448123,
|
|
"loss": 5.6913,
|
|
"mean_token_accuracy": 0.16798634380102156,
|
|
"num_tokens": 6188116.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.503647756576538,
|
|
"epoch": 2.9843158464034616,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00045127062244728353,
|
|
"loss": 5.4867,
|
|
"mean_token_accuracy": 0.18498665243387222,
|
|
"num_tokens": 6198606.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.656757736206055,
|
|
"epoch": 2.989724175229854,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004510037350819055,
|
|
"loss": 5.5214,
|
|
"mean_token_accuracy": 0.17884967774152755,
|
|
"num_tokens": 6210366.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.470056104660034,
|
|
"epoch": 2.9951325040562464,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045073620811854744,
|
|
"loss": 5.4044,
|
|
"mean_token_accuracy": 0.18297170102596283,
|
|
"num_tokens": 6221376.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.541294362809923,
|
|
"epoch": 3.0,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 0.00045046804252940294,
|
|
"loss": 5.3931,
|
|
"mean_token_accuracy": 0.18202034301227993,
|
|
"num_tokens": 6230667.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.502279281616211,
|
|
"epoch": 3.005408328826393,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045019923928898644,
|
|
"loss": 5.0513,
|
|
"mean_token_accuracy": 0.19762913584709169,
|
|
"num_tokens": 6241158.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.540598630905151,
|
|
"epoch": 3.0108166576527853,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004499297993741294,
|
|
"loss": 5.1571,
|
|
"mean_token_accuracy": 0.18719540536403656,
|
|
"num_tokens": 6253811.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.502381324768066,
|
|
"epoch": 3.016224986479178,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004496597237639773,
|
|
"loss": 5.1047,
|
|
"mean_token_accuracy": 0.19511241167783738,
|
|
"num_tokens": 6265066.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.414208793640137,
|
|
"epoch": 3.0216333153055706,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044938901343998536,
|
|
"loss": 5.0968,
|
|
"mean_token_accuracy": 0.19246386736631393,
|
|
"num_tokens": 6275889.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.460005617141723,
|
|
"epoch": 3.0270416441319634,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004491176693859156,
|
|
"loss": 5.0428,
|
|
"mean_token_accuracy": 0.20304037928581237,
|
|
"num_tokens": 6287576.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.438098526000976,
|
|
"epoch": 3.032449972958356,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004488456925878329,
|
|
"loss": 5.1274,
|
|
"mean_token_accuracy": 0.19501543492078782,
|
|
"num_tokens": 6299419.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.462873792648315,
|
|
"epoch": 3.0378583017847487,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044857308403410163,
|
|
"loss": 5.1135,
|
|
"mean_token_accuracy": 0.20509681552648545,
|
|
"num_tokens": 6310464.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.496682929992676,
|
|
"epoch": 3.043266630611141,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044829984471538183,
|
|
"loss": 5.1044,
|
|
"mean_token_accuracy": 0.19959417432546617,
|
|
"num_tokens": 6321442.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.476576805114746,
|
|
"epoch": 3.048674959437534,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004480259756246258,
|
|
"loss": 5.1568,
|
|
"mean_token_accuracy": 0.19579172432422637,
|
|
"num_tokens": 6332868.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.543281126022339,
|
|
"epoch": 3.0540832882639264,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044775147775707465,
|
|
"loss": 5.1499,
|
|
"mean_token_accuracy": 0.19697192907333375,
|
|
"num_tokens": 6345232.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.490166664123535,
|
|
"epoch": 3.0594916170903192,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000447476352110254,
|
|
"loss": 5.181,
|
|
"mean_token_accuracy": 0.19233484864234923,
|
|
"num_tokens": 6356524.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.491991662979126,
|
|
"epoch": 3.0648999459167117,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004472005996839712,
|
|
"loss": 5.1245,
|
|
"mean_token_accuracy": 0.19802421182394028,
|
|
"num_tokens": 6367590.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.499281167984009,
|
|
"epoch": 3.0703082747431045,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004469242214803112,
|
|
"loss": 5.1584,
|
|
"mean_token_accuracy": 0.18859623223543168,
|
|
"num_tokens": 6378211.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.44067759513855,
|
|
"epoch": 3.075716603569497,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004466472185036331,
|
|
"loss": 5.1604,
|
|
"mean_token_accuracy": 0.20044297277927398,
|
|
"num_tokens": 6389595.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.459574222564697,
|
|
"epoch": 3.08112493239589,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004463695917605663,
|
|
"loss": 5.1384,
|
|
"mean_token_accuracy": 0.19419959038496018,
|
|
"num_tokens": 6401286.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.4308734893798825,
|
|
"epoch": 3.086533261222282,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000446091342260007,
|
|
"loss": 5.1009,
|
|
"mean_token_accuracy": 0.201543128490448,
|
|
"num_tokens": 6411946.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.451775789260864,
|
|
"epoch": 3.091941590048675,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004458124710131145,
|
|
"loss": 5.0614,
|
|
"mean_token_accuracy": 0.20369820892810822,
|
|
"num_tokens": 6422463.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.335670900344849,
|
|
"epoch": 3.0973499188750675,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00044553297903330766,
|
|
"loss": 5.0618,
|
|
"mean_token_accuracy": 0.20009672343730928,
|
|
"num_tokens": 6433576.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.412241411209107,
|
|
"epoch": 3.1027582477014604,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044525286733626085,
|
|
"loss": 5.036,
|
|
"mean_token_accuracy": 0.20050158202648163,
|
|
"num_tokens": 6444058.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.450611686706543,
|
|
"epoch": 3.1081665765278528,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004449721369399008,
|
|
"loss": 5.2149,
|
|
"mean_token_accuracy": 0.19681848287582399,
|
|
"num_tokens": 6453872.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.428837680816651,
|
|
"epoch": 3.1135749053542456,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044469078886440227,
|
|
"loss": 5.1208,
|
|
"mean_token_accuracy": 0.1948903575539589,
|
|
"num_tokens": 6465367.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.415521192550659,
|
|
"epoch": 3.118983234180638,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044440882413218484,
|
|
"loss": 5.0636,
|
|
"mean_token_accuracy": 0.20829211175441742,
|
|
"num_tokens": 6476655.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.403928518295288,
|
|
"epoch": 3.124391563007031,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044412624376790905,
|
|
"loss": 5.1298,
|
|
"mean_token_accuracy": 0.19333246499300002,
|
|
"num_tokens": 6488348.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.465018224716187,
|
|
"epoch": 3.1297998918334233,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004438430487984726,
|
|
"loss": 5.1585,
|
|
"mean_token_accuracy": 0.19861571192741395,
|
|
"num_tokens": 6500127.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.339347171783447,
|
|
"epoch": 3.135208220659816,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004435592402530066,
|
|
"loss": 5.1164,
|
|
"mean_token_accuracy": 0.19745077043771744,
|
|
"num_tokens": 6512772.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.479548978805542,
|
|
"epoch": 3.1406165494862086,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000443274819162872,
|
|
"loss": 5.12,
|
|
"mean_token_accuracy": 0.20250731259584426,
|
|
"num_tokens": 6523359.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.36066164970398,
|
|
"epoch": 3.1460248783126015,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004429897865616557,
|
|
"loss": 5.1152,
|
|
"mean_token_accuracy": 0.19428427517414093,
|
|
"num_tokens": 6533679.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.4315262794494625,
|
|
"epoch": 3.151433207138994,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004427041434851668,
|
|
"loss": 5.0727,
|
|
"mean_token_accuracy": 0.19852081388235093,
|
|
"num_tokens": 6544694.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.384871530532837,
|
|
"epoch": 3.1568415359653867,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044241789097143293,
|
|
"loss": 5.0992,
|
|
"mean_token_accuracy": 0.1995680123567581,
|
|
"num_tokens": 6555858.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.432562446594238,
|
|
"epoch": 3.162249864791779,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00044213103006069635,
|
|
"loss": 5.1491,
|
|
"mean_token_accuracy": 0.19816354066133499,
|
|
"num_tokens": 6568214.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.338097715377808,
|
|
"epoch": 3.167658193618172,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044184356179541035,
|
|
"loss": 5.087,
|
|
"mean_token_accuracy": 0.19828221052885056,
|
|
"num_tokens": 6579646.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.446574306488037,
|
|
"epoch": 3.1730665224445644,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004415554872202352,
|
|
"loss": 5.0468,
|
|
"mean_token_accuracy": 0.20976133048534393,
|
|
"num_tokens": 6590683.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.424098539352417,
|
|
"epoch": 3.1784748512709573,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00044126680738203466,
|
|
"loss": 5.1838,
|
|
"mean_token_accuracy": 0.19457100927829743,
|
|
"num_tokens": 6601197.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.430385875701904,
|
|
"epoch": 3.1838831800973497,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004409775233298718,
|
|
"loss": 5.2447,
|
|
"mean_token_accuracy": 0.18377748876810074,
|
|
"num_tokens": 6613564.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.43729567527771,
|
|
"epoch": 3.1892915089237426,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004406876361150056,
|
|
"loss": 5.0959,
|
|
"mean_token_accuracy": 0.1958130657672882,
|
|
"num_tokens": 6624696.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.380353689193726,
|
|
"epoch": 3.194699837750135,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004403971467908869,
|
|
"loss": 5.1082,
|
|
"mean_token_accuracy": 0.2080030098557472,
|
|
"num_tokens": 6636302.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.455491065979004,
|
|
"epoch": 3.200108166576528,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004401060564131545,
|
|
"loss": 5.1433,
|
|
"mean_token_accuracy": 0.20008057206869126,
|
|
"num_tokens": 6646517.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.3872395038604735,
|
|
"epoch": 3.2055164954029207,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004398143660396315,
|
|
"loss": 5.1291,
|
|
"mean_token_accuracy": 0.19616604298353196,
|
|
"num_tokens": 6657768.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.3018372535705565,
|
|
"epoch": 3.210924824229313,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004395220767303215,
|
|
"loss": 5.047,
|
|
"mean_token_accuracy": 0.2003517523407936,
|
|
"num_tokens": 6669110.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.444350814819336,
|
|
"epoch": 3.2163331530557056,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004392291895474045,
|
|
"loss": 5.1312,
|
|
"mean_token_accuracy": 0.18839288949966432,
|
|
"num_tokens": 6680925.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.442668151855469,
|
|
"epoch": 3.2217414818820984,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004389357055552331,
|
|
"loss": 5.225,
|
|
"mean_token_accuracy": 0.1943958058953285,
|
|
"num_tokens": 6692923.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.480123615264892,
|
|
"epoch": 3.2271498107084913,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00043864162582032897,
|
|
"loss": 5.0875,
|
|
"mean_token_accuracy": 0.19728373885154724,
|
|
"num_tokens": 6702692.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.38284707069397,
|
|
"epoch": 3.2325581395348837,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004383469514113784,
|
|
"loss": 5.1766,
|
|
"mean_token_accuracy": 0.19420597404241563,
|
|
"num_tokens": 6713830.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.482458877563476,
|
|
"epoch": 3.2379664683612766,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004380516833992289,
|
|
"loss": 5.162,
|
|
"mean_token_accuracy": 0.19368472695350647,
|
|
"num_tokens": 6725564.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.390593433380127,
|
|
"epoch": 3.243374797187669,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000437755822856885,
|
|
"loss": 5.1216,
|
|
"mean_token_accuracy": 0.19481516480445862,
|
|
"num_tokens": 6736563.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.243374797187669,
|
|
"eval_entropy": 5.306998636505821,
|
|
"eval_loss": 5.960112571716309,
|
|
"eval_mean_token_accuracy": 0.1720793664116751,
|
|
"eval_num_tokens": 6736563.0,
|
|
"eval_runtime": 2.5386,
|
|
"eval_samples_per_second": 1384.624,
|
|
"eval_steps_per_second": 173.324,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.3262331008911135,
|
|
"epoch": 3.248783126014062,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004374593708595049,
|
|
"loss": 5.0858,
|
|
"mean_token_accuracy": 0.19832624942064286,
|
|
"num_tokens": 6745845.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.346648693084717,
|
|
"epoch": 3.2541914548404542,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00043716232848439555,
|
|
"loss": 5.0537,
|
|
"mean_token_accuracy": 0.200261852145195,
|
|
"num_tokens": 6757450.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.44229645729065,
|
|
"epoch": 3.259599783666847,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00043686469681100985,
|
|
"loss": 5.2145,
|
|
"mean_token_accuracy": 0.1953560322523117,
|
|
"num_tokens": 6768542.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.379146003723145,
|
|
"epoch": 3.2650081124932395,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00043656647692094186,
|
|
"loss": 5.1294,
|
|
"mean_token_accuracy": 0.20932810604572297,
|
|
"num_tokens": 6779873.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.375614404678345,
|
|
"epoch": 3.2704164413196324,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004362676698979233,
|
|
"loss": 5.1704,
|
|
"mean_token_accuracy": 0.20028591454029082,
|
|
"num_tokens": 6790898.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.423163890838623,
|
|
"epoch": 3.275824770146025,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043596827682781974,
|
|
"loss": 5.1372,
|
|
"mean_token_accuracy": 0.1960364669561386,
|
|
"num_tokens": 6802175.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.467242431640625,
|
|
"epoch": 3.2812330989724177,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004356682987986262,
|
|
"loss": 5.2658,
|
|
"mean_token_accuracy": 0.1908641755580902,
|
|
"num_tokens": 6813675.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.387868118286133,
|
|
"epoch": 3.28664142779881,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004353677369004635,
|
|
"loss": 5.1168,
|
|
"mean_token_accuracy": 0.1980368375778198,
|
|
"num_tokens": 6825020.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.420724582672119,
|
|
"epoch": 3.292049756625203,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043506659222557426,
|
|
"loss": 5.0141,
|
|
"mean_token_accuracy": 0.21363207697868347,
|
|
"num_tokens": 6834954.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.446198844909668,
|
|
"epoch": 3.2974580854515954,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000434764865868319,
|
|
"loss": 5.262,
|
|
"mean_token_accuracy": 0.18776430636644365,
|
|
"num_tokens": 6847958.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.41791844367981,
|
|
"epoch": 3.302866414277988,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000434462558925172,
|
|
"loss": 5.0701,
|
|
"mean_token_accuracy": 0.20969656854867935,
|
|
"num_tokens": 6858782.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.261764669418335,
|
|
"epoch": 3.3082747431043806,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043415967249471734,
|
|
"loss": 5.0799,
|
|
"mean_token_accuracy": 0.20303125977516173,
|
|
"num_tokens": 6869807.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.452825117111206,
|
|
"epoch": 3.3136830719307735,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000433856207677645,
|
|
"loss": 5.1309,
|
|
"mean_token_accuracy": 0.19971420913934707,
|
|
"num_tokens": 6880642.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.4816100120544435,
|
|
"epoch": 3.319091400757166,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000433552165576747,
|
|
"loss": 5.2251,
|
|
"mean_token_accuracy": 0.18565797805786133,
|
|
"num_tokens": 6891677.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.380715703964233,
|
|
"epoch": 3.3244997295835588,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00043324754729691275,
|
|
"loss": 5.1238,
|
|
"mean_token_accuracy": 0.19887210428714752,
|
|
"num_tokens": 6903383.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.467216634750367,
|
|
"epoch": 3.329908058409951,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043294235394512594,
|
|
"loss": 5.1863,
|
|
"mean_token_accuracy": 0.20602514445781708,
|
|
"num_tokens": 6915874.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.315272665023803,
|
|
"epoch": 3.335316387236344,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004326365866304599,
|
|
"loss": 4.9979,
|
|
"mean_token_accuracy": 0.20175593346357346,
|
|
"num_tokens": 6926356.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.4103374004364015,
|
|
"epoch": 3.3407247160627365,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004323302464640738,
|
|
"loss": 5.1357,
|
|
"mean_token_accuracy": 0.20278737246990203,
|
|
"num_tokens": 6937678.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.384547901153565,
|
|
"epoch": 3.3461330448891293,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043202333455920843,
|
|
"loss": 5.1404,
|
|
"mean_token_accuracy": 0.19868416488170623,
|
|
"num_tokens": 6949447.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.388951683044434,
|
|
"epoch": 3.3515413737155217,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004317158520311824,
|
|
"loss": 5.0588,
|
|
"mean_token_accuracy": 0.19861841052770615,
|
|
"num_tokens": 6960347.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.319481468200683,
|
|
"epoch": 3.3569497025419146,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004314077999973879,
|
|
"loss": 5.0725,
|
|
"mean_token_accuracy": 0.2026567280292511,
|
|
"num_tokens": 6971353.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.397992849349976,
|
|
"epoch": 3.362358031368307,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004310991795772869,
|
|
"loss": 5.2052,
|
|
"mean_token_accuracy": 0.18833321332931519,
|
|
"num_tokens": 6982519.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.33483476638794,
|
|
"epoch": 3.3677663601947,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004307899918924065,
|
|
"loss": 5.0616,
|
|
"mean_token_accuracy": 0.20603248924016954,
|
|
"num_tokens": 6994021.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.343628406524658,
|
|
"epoch": 3.3731746890210923,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004304802380663356,
|
|
"loss": 5.1576,
|
|
"mean_token_accuracy": 0.2018789380788803,
|
|
"num_tokens": 7005505.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.4013995170593265,
|
|
"epoch": 3.378583017847485,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00043016991922472015,
|
|
"loss": 5.1944,
|
|
"mean_token_accuracy": 0.19918280690908433,
|
|
"num_tokens": 7016123.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.465548992156982,
|
|
"epoch": 3.383991346673878,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00042985903649525977,
|
|
"loss": 5.1336,
|
|
"mean_token_accuracy": 0.1994956001639366,
|
|
"num_tokens": 7027083.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.3966175556182865,
|
|
"epoch": 3.3893996755002704,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00042954759100770275,
|
|
"loss": 5.2179,
|
|
"mean_token_accuracy": 0.19297376722097398,
|
|
"num_tokens": 7037892.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.400795602798462,
|
|
"epoch": 3.394808004326663,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042923558389384303,
|
|
"loss": 5.1868,
|
|
"mean_token_accuracy": 0.19257848411798478,
|
|
"num_tokens": 7048996.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.462365055084229,
|
|
"epoch": 3.4002163331530557,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004289230162875149,
|
|
"loss": 5.219,
|
|
"mean_token_accuracy": 0.1920529469847679,
|
|
"num_tokens": 7060884.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.427655649185181,
|
|
"epoch": 3.4056246619794486,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042860988932458987,
|
|
"loss": 5.2091,
|
|
"mean_token_accuracy": 0.19274180829524995,
|
|
"num_tokens": 7072550.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.340734148025513,
|
|
"epoch": 3.411032990805841,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004282962041429721,
|
|
"loss": 5.1364,
|
|
"mean_token_accuracy": 0.20432330369949342,
|
|
"num_tokens": 7085325.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.318571949005127,
|
|
"epoch": 3.4164413196322334,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00042798196188259427,
|
|
"loss": 5.0112,
|
|
"mean_token_accuracy": 0.19978462755680085,
|
|
"num_tokens": 7095691.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.235277843475342,
|
|
"epoch": 3.4218496484586263,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004276671636854133,
|
|
"loss": 4.987,
|
|
"mean_token_accuracy": 0.2108358308672905,
|
|
"num_tokens": 7106552.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.270534467697144,
|
|
"epoch": 3.427257977285019,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042735181069540677,
|
|
"loss": 5.1295,
|
|
"mean_token_accuracy": 0.19305755198001862,
|
|
"num_tokens": 7119708.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.396907615661621,
|
|
"epoch": 3.4326663061114115,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004270359040585681,
|
|
"loss": 5.1231,
|
|
"mean_token_accuracy": 0.19761836379766465,
|
|
"num_tokens": 7130663.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.279340219497681,
|
|
"epoch": 3.4380746349378044,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00042671944492290277,
|
|
"loss": 5.0603,
|
|
"mean_token_accuracy": 0.1948313072323799,
|
|
"num_tokens": 7141513.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.359659719467163,
|
|
"epoch": 3.443482963764197,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004264024344384239,
|
|
"loss": 5.1306,
|
|
"mean_token_accuracy": 0.19705723971128464,
|
|
"num_tokens": 7153387.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.368605661392212,
|
|
"epoch": 3.4488912925905897,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042608487375714847,
|
|
"loss": 5.1376,
|
|
"mean_token_accuracy": 0.1991734027862549,
|
|
"num_tokens": 7164905.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.360475683212281,
|
|
"epoch": 3.454299621416982,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00042576676403309265,
|
|
"loss": 5.1627,
|
|
"mean_token_accuracy": 0.2047038644552231,
|
|
"num_tokens": 7175636.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.392428779602051,
|
|
"epoch": 3.459707950243375,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004254481064222679,
|
|
"loss": 5.1453,
|
|
"mean_token_accuracy": 0.1974661722779274,
|
|
"num_tokens": 7185903.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.29183259010315,
|
|
"epoch": 3.4651162790697674,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00042512890208267677,
|
|
"loss": 5.0788,
|
|
"mean_token_accuracy": 0.1953657701611519,
|
|
"num_tokens": 7198257.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.420766115188599,
|
|
"epoch": 3.4705246078961602,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042480915217430853,
|
|
"loss": 5.0483,
|
|
"mean_token_accuracy": 0.2114559069275856,
|
|
"num_tokens": 7208839.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.380645322799682,
|
|
"epoch": 3.4759329367225527,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00042448885785913507,
|
|
"loss": 5.1787,
|
|
"mean_token_accuracy": 0.1936534345149994,
|
|
"num_tokens": 7221078.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.3577696800231935,
|
|
"epoch": 3.4813412655489455,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00042416802030110677,
|
|
"loss": 5.078,
|
|
"mean_token_accuracy": 0.200352181494236,
|
|
"num_tokens": 7230287.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.33955659866333,
|
|
"epoch": 3.486749594375338,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042384664066614787,
|
|
"loss": 5.1655,
|
|
"mean_token_accuracy": 0.19922966957092286,
|
|
"num_tokens": 7242253.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.399269247055054,
|
|
"epoch": 3.492157923201731,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004235247201221528,
|
|
"loss": 5.1324,
|
|
"mean_token_accuracy": 0.20308125466108323,
|
|
"num_tokens": 7254098.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.215890884399414,
|
|
"epoch": 3.497566252028123,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004232022598389816,
|
|
"loss": 5.035,
|
|
"mean_token_accuracy": 0.20540016740560532,
|
|
"num_tokens": 7264153.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.311744451522827,
|
|
"epoch": 3.502974580854516,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004228792609884557,
|
|
"loss": 5.1645,
|
|
"mean_token_accuracy": 0.19657052010297776,
|
|
"num_tokens": 7276431.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.402192258834839,
|
|
"epoch": 3.5083829096809085,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00042255572474435364,
|
|
"loss": 5.099,
|
|
"mean_token_accuracy": 0.2001231402158737,
|
|
"num_tokens": 7287882.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.336764526367188,
|
|
"epoch": 3.5137912385073014,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004222316522824069,
|
|
"loss": 5.229,
|
|
"mean_token_accuracy": 0.1924380749464035,
|
|
"num_tokens": 7298846.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.402611494064331,
|
|
"epoch": 3.5191995673336938,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004219070447802955,
|
|
"loss": 5.122,
|
|
"mean_token_accuracy": 0.19636227935552597,
|
|
"num_tokens": 7310372.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.445440864562988,
|
|
"epoch": 3.5246078961600866,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042158190341764384,
|
|
"loss": 5.1943,
|
|
"mean_token_accuracy": 0.19431152939796448,
|
|
"num_tokens": 7322200.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.323125267028809,
|
|
"epoch": 3.530016224986479,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00042125622937601645,
|
|
"loss": 5.0749,
|
|
"mean_token_accuracy": 0.20254762023687362,
|
|
"num_tokens": 7331824.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.322934579849243,
|
|
"epoch": 3.535424553812872,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004209300238389135,
|
|
"loss": 5.0695,
|
|
"mean_token_accuracy": 0.20224796682596208,
|
|
"num_tokens": 7341522.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.3828519821167,
|
|
"epoch": 3.5408328826392643,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004206032879917667,
|
|
"loss": 5.153,
|
|
"mean_token_accuracy": 0.20568514317274095,
|
|
"num_tokens": 7353312.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.383207178115844,
|
|
"epoch": 3.546241211465657,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004202760230219348,
|
|
"loss": 5.1455,
|
|
"mean_token_accuracy": 0.209283047914505,
|
|
"num_tokens": 7363985.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.344035482406616,
|
|
"epoch": 3.5516495402920496,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00041994823011869955,
|
|
"loss": 5.1292,
|
|
"mean_token_accuracy": 0.1983759194612503,
|
|
"num_tokens": 7373872.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.407253551483154,
|
|
"epoch": 3.5570578691184425,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00041961991047326116,
|
|
"loss": 5.1553,
|
|
"mean_token_accuracy": 0.20117487460374833,
|
|
"num_tokens": 7386759.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.3192907810211185,
|
|
"epoch": 3.5624661979448353,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004192910652787338,
|
|
"loss": 5.0946,
|
|
"mean_token_accuracy": 0.2043826773762703,
|
|
"num_tokens": 7398001.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.42265305519104,
|
|
"epoch": 3.5678745267712277,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004189616957301419,
|
|
"loss": 5.1711,
|
|
"mean_token_accuracy": 0.19592228829860686,
|
|
"num_tokens": 7409159.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.313607549667358,
|
|
"epoch": 3.57328285559762,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004186318030244149,
|
|
"loss": 5.0936,
|
|
"mean_token_accuracy": 0.20098270922899247,
|
|
"num_tokens": 7421028.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.337180948257446,
|
|
"epoch": 3.578691184424013,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041830138836038383,
|
|
"loss": 5.1896,
|
|
"mean_token_accuracy": 0.19479376822710037,
|
|
"num_tokens": 7432321.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.438666439056396,
|
|
"epoch": 3.584099513250406,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004179704529387763,
|
|
"loss": 5.2071,
|
|
"mean_token_accuracy": 0.20270422399044036,
|
|
"num_tokens": 7443607.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.36533670425415,
|
|
"epoch": 3.5895078420767983,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004176389979622125,
|
|
"loss": 5.0717,
|
|
"mean_token_accuracy": 0.20045875757932663,
|
|
"num_tokens": 7455222.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.36827335357666,
|
|
"epoch": 3.5949161709031907,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004173070246352003,
|
|
"loss": 5.2365,
|
|
"mean_token_accuracy": 0.19650563597679138,
|
|
"num_tokens": 7468511.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.397350406646728,
|
|
"epoch": 3.6003244997295836,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00041697453416413184,
|
|
"loss": 5.2109,
|
|
"mean_token_accuracy": 0.20041104257106782,
|
|
"num_tokens": 7479810.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.2944831371307375,
|
|
"epoch": 3.6057328285559764,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000416641527757278,
|
|
"loss": 5.0477,
|
|
"mean_token_accuracy": 0.19987694919109344,
|
|
"num_tokens": 7491405.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.371123170852661,
|
|
"epoch": 3.611141157382369,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00041630800662478514,
|
|
"loss": 5.1616,
|
|
"mean_token_accuracy": 0.19249594062566758,
|
|
"num_tokens": 7501820.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.386195850372315,
|
|
"epoch": 3.6165494862087613,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00041597397197866957,
|
|
"loss": 5.1299,
|
|
"mean_token_accuracy": 0.19731049090623856,
|
|
"num_tokens": 7512165.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.280309057235717,
|
|
"epoch": 3.621957815035154,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004156394250328141,
|
|
"loss": 5.1168,
|
|
"mean_token_accuracy": 0.20301333218812942,
|
|
"num_tokens": 7523755.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.31842942237854,
|
|
"epoch": 3.627366143861547,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000415304367002963,
|
|
"loss": 5.1135,
|
|
"mean_token_accuracy": 0.19609812200069426,
|
|
"num_tokens": 7535674.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.346430683135987,
|
|
"epoch": 3.6327744726879394,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00041496879910671796,
|
|
"loss": 5.103,
|
|
"mean_token_accuracy": 0.20175676345825194,
|
|
"num_tokens": 7546885.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.28568205833435,
|
|
"epoch": 3.638182801514332,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004146327225635336,
|
|
"loss": 5.0868,
|
|
"mean_token_accuracy": 0.2036902964115143,
|
|
"num_tokens": 7557718.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.2606817245483395,
|
|
"epoch": 3.6435911303407247,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004142961385947127,
|
|
"loss": 5.0766,
|
|
"mean_token_accuracy": 0.1953711360692978,
|
|
"num_tokens": 7569472.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.413916158676147,
|
|
"epoch": 3.6489994591671175,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00041395904842340215,
|
|
"loss": 5.0962,
|
|
"mean_token_accuracy": 0.1925591766834259,
|
|
"num_tokens": 7579954.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.31058931350708,
|
|
"epoch": 3.65440778799351,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004136214532745887,
|
|
"loss": 5.041,
|
|
"mean_token_accuracy": 0.20725557208061218,
|
|
"num_tokens": 7590717.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.418099975585937,
|
|
"epoch": 3.6598161168199024,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004132833543750936,
|
|
"loss": 5.3702,
|
|
"mean_token_accuracy": 0.18772991448640824,
|
|
"num_tokens": 7602802.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.491369533538818,
|
|
"epoch": 3.6652244456462952,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00041294475295356917,
|
|
"loss": 5.2899,
|
|
"mean_token_accuracy": 0.19091144502162932,
|
|
"num_tokens": 7614944.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.3593004703521725,
|
|
"epoch": 3.670632774472688,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004126056502404937,
|
|
"loss": 5.1798,
|
|
"mean_token_accuracy": 0.1929037481546402,
|
|
"num_tokens": 7626449.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.314221143722534,
|
|
"epoch": 3.6760411032990805,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004122660474681672,
|
|
"loss": 5.0629,
|
|
"mean_token_accuracy": 0.20498253405094147,
|
|
"num_tokens": 7637878.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.427420902252197,
|
|
"epoch": 3.6814494321254734,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00041192594587070706,
|
|
"loss": 5.2794,
|
|
"mean_token_accuracy": 0.188936085999012,
|
|
"num_tokens": 7649428.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.452918148040771,
|
|
"epoch": 3.686857760951866,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00041158534668404325,
|
|
"loss": 5.1941,
|
|
"mean_token_accuracy": 0.19379565268754959,
|
|
"num_tokens": 7660527.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.282226276397705,
|
|
"epoch": 3.6922660897782587,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00041124425114591386,
|
|
"loss": 5.1092,
|
|
"mean_token_accuracy": 0.20502447634935378,
|
|
"num_tokens": 7671249.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.401538610458374,
|
|
"epoch": 3.697674418604651,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00041090266049586107,
|
|
"loss": 5.1144,
|
|
"mean_token_accuracy": 0.19788116216659546,
|
|
"num_tokens": 7681824.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.3221875667572025,
|
|
"epoch": 3.703082747431044,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004105605759752258,
|
|
"loss": 5.17,
|
|
"mean_token_accuracy": 0.19329199939966202,
|
|
"num_tokens": 7691916.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.4061953067779545,
|
|
"epoch": 3.7084910762574363,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00041021799882714436,
|
|
"loss": 5.2093,
|
|
"mean_token_accuracy": 0.19577225893735886,
|
|
"num_tokens": 7703355.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.3635279655456545,
|
|
"epoch": 3.713899405083829,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004098749302965426,
|
|
"loss": 5.1472,
|
|
"mean_token_accuracy": 0.20143863558769226,
|
|
"num_tokens": 7714731.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.3299795627594,
|
|
"epoch": 3.7193077339102216,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004095313716301324,
|
|
"loss": 5.0838,
|
|
"mean_token_accuracy": 0.20690459609031678,
|
|
"num_tokens": 7725696.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.31907205581665,
|
|
"epoch": 3.7247160627366145,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004091873240764069,
|
|
"loss": 5.1019,
|
|
"mean_token_accuracy": 0.19736230373382568,
|
|
"num_tokens": 7737121.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.429636526107788,
|
|
"epoch": 3.730124391563007,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00040884278888563566,
|
|
"loss": 5.2084,
|
|
"mean_token_accuracy": 0.19627318084239959,
|
|
"num_tokens": 7749088.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.429543924331665,
|
|
"epoch": 3.7355327203893998,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00040849776730986014,
|
|
"loss": 5.2972,
|
|
"mean_token_accuracy": 0.19169671833515167,
|
|
"num_tokens": 7760676.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.385066652297974,
|
|
"epoch": 3.740941049215792,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004081522606028899,
|
|
"loss": 5.2044,
|
|
"mean_token_accuracy": 0.19767188131809235,
|
|
"num_tokens": 7772560.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.3613731384277346,
|
|
"epoch": 3.746349378042185,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00040780627002029674,
|
|
"loss": 5.1168,
|
|
"mean_token_accuracy": 0.20275183618068696,
|
|
"num_tokens": 7782206.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.414861965179443,
|
|
"epoch": 3.7517577068685775,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004074597968194116,
|
|
"loss": 5.2326,
|
|
"mean_token_accuracy": 0.18894226849079132,
|
|
"num_tokens": 7793862.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.393022251129151,
|
|
"epoch": 3.7571660356949703,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004071128422593186,
|
|
"loss": 5.255,
|
|
"mean_token_accuracy": 0.18937607556581498,
|
|
"num_tokens": 7805291.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.461092948913574,
|
|
"epoch": 3.7625743645213627,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004067654076008515,
|
|
"loss": 5.2173,
|
|
"mean_token_accuracy": 0.19551273733377456,
|
|
"num_tokens": 7816141.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.350432252883911,
|
|
"epoch": 3.7679826933477556,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00040641749410658877,
|
|
"loss": 5.1891,
|
|
"mean_token_accuracy": 0.18398540914058686,
|
|
"num_tokens": 7826209.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.3595129489898685,
|
|
"epoch": 3.773391022174148,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004060691030408487,
|
|
"loss": 5.1436,
|
|
"mean_token_accuracy": 0.19257761985063554,
|
|
"num_tokens": 7836734.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.384549999237061,
|
|
"epoch": 3.778799351000541,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00040572023566968536,
|
|
"loss": 5.2079,
|
|
"mean_token_accuracy": 0.19411968439817429,
|
|
"num_tokens": 7848283.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.4173033237457275,
|
|
"epoch": 3.7842076798269337,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00040537089326088345,
|
|
"loss": 5.1048,
|
|
"mean_token_accuracy": 0.19854387491941453,
|
|
"num_tokens": 7861226.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.254236268997192,
|
|
"epoch": 3.789616008653326,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004050210770839541,
|
|
"loss": 5.0808,
|
|
"mean_token_accuracy": 0.2079720288515091,
|
|
"num_tokens": 7872426.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.363286018371582,
|
|
"epoch": 3.7950243374797186,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004046707884101303,
|
|
"loss": 5.1657,
|
|
"mean_token_accuracy": 0.195390984416008,
|
|
"num_tokens": 7883884.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.31299204826355,
|
|
"epoch": 3.8004326663061114,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040432002851236164,
|
|
"loss": 5.1021,
|
|
"mean_token_accuracy": 0.20301491618156434,
|
|
"num_tokens": 7895415.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.327358436584473,
|
|
"epoch": 3.8058409951325043,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004039687986653106,
|
|
"loss": 5.2006,
|
|
"mean_token_accuracy": 0.19163492769002916,
|
|
"num_tokens": 7907444.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.484244871139526,
|
|
"epoch": 3.8112493239588967,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00040361710014534714,
|
|
"loss": 5.1786,
|
|
"mean_token_accuracy": 0.19134418815374374,
|
|
"num_tokens": 7918811.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.395617818832397,
|
|
"epoch": 3.816657652785289,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004032649342305445,
|
|
"loss": 5.2493,
|
|
"mean_token_accuracy": 0.18802796751260759,
|
|
"num_tokens": 7930657.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.2587381362915036,
|
|
"epoch": 3.822065981611682,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004029123022006745,
|
|
"loss": 5.0864,
|
|
"mean_token_accuracy": 0.19073489904403687,
|
|
"num_tokens": 7942469.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.404836797714234,
|
|
"epoch": 3.827474310438075,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004025592053372027,
|
|
"loss": 5.2442,
|
|
"mean_token_accuracy": 0.19406743794679643,
|
|
"num_tokens": 7954034.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.357091045379638,
|
|
"epoch": 3.8328826392644673,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004022056449232839,
|
|
"loss": 5.1505,
|
|
"mean_token_accuracy": 0.1996315374970436,
|
|
"num_tokens": 7966079.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.23913893699646,
|
|
"epoch": 3.8382909680908597,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004018516222437575,
|
|
"loss": 5.1176,
|
|
"mean_token_accuracy": 0.2036358118057251,
|
|
"num_tokens": 7977785.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.471898984909058,
|
|
"epoch": 3.8436992969172525,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004014971385851428,
|
|
"loss": 5.2741,
|
|
"mean_token_accuracy": 0.19076469242572786,
|
|
"num_tokens": 7988874.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.333814668655395,
|
|
"epoch": 3.8491076257436454,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000401142195235634,
|
|
"loss": 5.0546,
|
|
"mean_token_accuracy": 0.20550299882888795,
|
|
"num_tokens": 7999438.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.321576261520386,
|
|
"epoch": 3.854515954570038,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004007867934850963,
|
|
"loss": 5.1452,
|
|
"mean_token_accuracy": 0.20005037784576415,
|
|
"num_tokens": 8009658.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.379993343353272,
|
|
"epoch": 3.8599242833964302,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004004309346250603,
|
|
"loss": 5.1193,
|
|
"mean_token_accuracy": 0.19628584235906602,
|
|
"num_tokens": 8020988.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.30539903640747,
|
|
"epoch": 3.865332612222823,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004000746199487181,
|
|
"loss": 5.005,
|
|
"mean_token_accuracy": 0.2062746524810791,
|
|
"num_tokens": 8031329.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.233512830734253,
|
|
"epoch": 3.870740941049216,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000399717850750918,
|
|
"loss": 5.1102,
|
|
"mean_token_accuracy": 0.19695514738559722,
|
|
"num_tokens": 8041693.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.37098536491394,
|
|
"epoch": 3.8761492698756084,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039936062832816,
|
|
"loss": 5.162,
|
|
"mean_token_accuracy": 0.20135014802217482,
|
|
"num_tokens": 8053123.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.394336271286011,
|
|
"epoch": 3.8815575987020012,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003990029539785913,
|
|
"loss": 5.1606,
|
|
"mean_token_accuracy": 0.196461521089077,
|
|
"num_tokens": 8064270.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.342864274978638,
|
|
"epoch": 3.8869659275283936,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039864482900200135,
|
|
"loss": 5.0796,
|
|
"mean_token_accuracy": 0.20653983503580092,
|
|
"num_tokens": 8074631.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.3207845211029055,
|
|
"epoch": 3.8923742563547865,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003982862546998169,
|
|
"loss": 5.1533,
|
|
"mean_token_accuracy": 0.20086930096149444,
|
|
"num_tokens": 8086794.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.370695495605469,
|
|
"epoch": 3.897782585181179,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003979272323750981,
|
|
"loss": 5.1599,
|
|
"mean_token_accuracy": 0.2002336710691452,
|
|
"num_tokens": 8097595.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.3372156620025635,
|
|
"epoch": 3.903190914007572,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003975677633325327,
|
|
"loss": 5.1316,
|
|
"mean_token_accuracy": 0.199468794465065,
|
|
"num_tokens": 8107174.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.317599773406982,
|
|
"epoch": 3.908599242833964,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003972078488784321,
|
|
"loss": 5.1421,
|
|
"mean_token_accuracy": 0.20067692846059798,
|
|
"num_tokens": 8118065.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.276710319519043,
|
|
"epoch": 3.914007571660357,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003968474903207263,
|
|
"loss": 5.1233,
|
|
"mean_token_accuracy": 0.19370758682489395,
|
|
"num_tokens": 8130604.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.367471313476562,
|
|
"epoch": 3.9194159004867495,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003964866889689592,
|
|
"loss": 5.1809,
|
|
"mean_token_accuracy": 0.19862312823534012,
|
|
"num_tokens": 8141808.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.4145020008087155,
|
|
"epoch": 3.9248242293131423,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003961254461342834,
|
|
"loss": 5.1562,
|
|
"mean_token_accuracy": 0.19601356387138366,
|
|
"num_tokens": 8152101.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.376813840866089,
|
|
"epoch": 3.9302325581395348,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003957637631294566,
|
|
"loss": 5.1817,
|
|
"mean_token_accuracy": 0.19523661434650422,
|
|
"num_tokens": 8163963.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.360791969299316,
|
|
"epoch": 3.9356408869659276,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039540164126883554,
|
|
"loss": 5.0998,
|
|
"mean_token_accuracy": 0.20471465587615967,
|
|
"num_tokens": 8174063.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.295090627670288,
|
|
"epoch": 3.94104921579232,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003950390818683718,
|
|
"loss": 5.1036,
|
|
"mean_token_accuracy": 0.20145625472068787,
|
|
"num_tokens": 8186800.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.341031265258789,
|
|
"epoch": 3.946457544618713,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003946760862456071,
|
|
"loss": 5.1726,
|
|
"mean_token_accuracy": 0.19883247017860411,
|
|
"num_tokens": 8198420.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.461062097549439,
|
|
"epoch": 3.9518658734451053,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003943126557196685,
|
|
"loss": 5.1998,
|
|
"mean_token_accuracy": 0.18966050893068315,
|
|
"num_tokens": 8209384.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.335430049896241,
|
|
"epoch": 3.957274202271498,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00039394879161126333,
|
|
"loss": 5.1604,
|
|
"mean_token_accuracy": 0.19565496891736983,
|
|
"num_tokens": 8220356.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.270952463150024,
|
|
"epoch": 3.9626825310978906,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003935844952426745,
|
|
"loss": 5.0979,
|
|
"mean_token_accuracy": 0.1977401003241539,
|
|
"num_tokens": 8231597.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.294128513336181,
|
|
"epoch": 3.9680908599242835,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00039321976793775586,
|
|
"loss": 5.0723,
|
|
"mean_token_accuracy": 0.20218011885881423,
|
|
"num_tokens": 8242554.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.400357866287232,
|
|
"epoch": 3.973499188750676,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003928546110219275,
|
|
"loss": 5.3019,
|
|
"mean_token_accuracy": 0.1910625860095024,
|
|
"num_tokens": 8253396.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.297176313400269,
|
|
"epoch": 3.9789075175770687,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039248902582217024,
|
|
"loss": 5.0758,
|
|
"mean_token_accuracy": 0.20564910620450974,
|
|
"num_tokens": 8265193.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.293642902374268,
|
|
"epoch": 3.9843158464034616,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00039212301366702153,
|
|
"loss": 5.1835,
|
|
"mean_token_accuracy": 0.199183851480484,
|
|
"num_tokens": 8276343.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.4490021705627445,
|
|
"epoch": 3.989724175229854,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039175657588657063,
|
|
"loss": 5.1669,
|
|
"mean_token_accuracy": 0.20016542822122574,
|
|
"num_tokens": 8288069.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.286073017120361,
|
|
"epoch": 3.9951325040562464,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003913897138124531,
|
|
"loss": 5.1299,
|
|
"mean_token_accuracy": 0.19696886837482452,
|
|
"num_tokens": 8299382.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.313399155934651,
|
|
"epoch": 4.0,
|
|
"grad_norm": 2.59375,
|
|
"learning_rate": 0.0003910224287778466,
|
|
"loss": 5.059,
|
|
"mean_token_accuracy": 0.20543919834825727,
|
|
"num_tokens": 8307556.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.3537312030792235,
|
|
"epoch": 4.005408328826393,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00039065472211746586,
|
|
"loss": 4.8608,
|
|
"mean_token_accuracy": 0.21980725973844528,
|
|
"num_tokens": 8317851.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.193052148818969,
|
|
"epoch": 4.010816657652786,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003902865951675576,
|
|
"loss": 4.7616,
|
|
"mean_token_accuracy": 0.22571386098861695,
|
|
"num_tokens": 8328485.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.3045131206512455,
|
|
"epoch": 4.016224986479178,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00038991804926589616,
|
|
"loss": 4.7339,
|
|
"mean_token_accuracy": 0.22240492403507234,
|
|
"num_tokens": 8339625.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.306188631057739,
|
|
"epoch": 4.0216333153055706,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000389549085751778,
|
|
"loss": 4.7966,
|
|
"mean_token_accuracy": 0.21777597814798355,
|
|
"num_tokens": 8350667.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.299514865875244,
|
|
"epoch": 4.027041644131963,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00038917970596601765,
|
|
"loss": 4.8402,
|
|
"mean_token_accuracy": 0.21808336526155472,
|
|
"num_tokens": 8362521.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.282034778594971,
|
|
"epoch": 4.032449972958356,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003888099112509419,
|
|
"loss": 4.9062,
|
|
"mean_token_accuracy": 0.20888912975788115,
|
|
"num_tokens": 8372703.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.401778316497802,
|
|
"epoch": 4.037858301784748,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00038843970295038564,
|
|
"loss": 4.8625,
|
|
"mean_token_accuracy": 0.21430229395627975,
|
|
"num_tokens": 8383345.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.210376739501953,
|
|
"epoch": 4.043266630611141,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00038806908240968666,
|
|
"loss": 4.7885,
|
|
"mean_token_accuracy": 0.21905164122581483,
|
|
"num_tokens": 8394056.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.234530782699585,
|
|
"epoch": 4.048674959437534,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000387698050975681,
|
|
"loss": 4.8195,
|
|
"mean_token_accuracy": 0.21632389426231385,
|
|
"num_tokens": 8403889.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.255437898635864,
|
|
"epoch": 4.054083288263927,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003873266099966976,
|
|
"loss": 4.8462,
|
|
"mean_token_accuracy": 0.21228063404560088,
|
|
"num_tokens": 8415169.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.275872278213501,
|
|
"epoch": 4.059491617090319,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038695476082255395,
|
|
"loss": 4.8656,
|
|
"mean_token_accuracy": 0.2096945270895958,
|
|
"num_tokens": 8427056.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.246340370178222,
|
|
"epoch": 4.064899945916712,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003865825048045507,
|
|
"loss": 4.8039,
|
|
"mean_token_accuracy": 0.21953081488609313,
|
|
"num_tokens": 8438900.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.270008277893067,
|
|
"epoch": 4.0703082747431045,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003862098432954672,
|
|
"loss": 4.8057,
|
|
"mean_token_accuracy": 0.21926842331886293,
|
|
"num_tokens": 8449037.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.189347982406616,
|
|
"epoch": 4.075716603569497,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003858367776495561,
|
|
"loss": 4.785,
|
|
"mean_token_accuracy": 0.21291797012090682,
|
|
"num_tokens": 8459409.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.30788950920105,
|
|
"epoch": 4.081124932395889,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00038546330922253886,
|
|
"loss": 4.8775,
|
|
"mean_token_accuracy": 0.21392126083374025,
|
|
"num_tokens": 8470190.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.316929578781128,
|
|
"epoch": 4.086533261222282,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003850894393716007,
|
|
"loss": 4.8706,
|
|
"mean_token_accuracy": 0.21057044416666032,
|
|
"num_tokens": 8481559.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.3330904006958,
|
|
"epoch": 4.091941590048675,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003847151694553855,
|
|
"loss": 4.8917,
|
|
"mean_token_accuracy": 0.20514337420463563,
|
|
"num_tokens": 8492607.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.311561059951782,
|
|
"epoch": 4.097349918875068,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003843405008339908,
|
|
"loss": 4.9905,
|
|
"mean_token_accuracy": 0.2053788959980011,
|
|
"num_tokens": 8504954.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.377383184432984,
|
|
"epoch": 4.10275824770146,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00038396543486896354,
|
|
"loss": 4.8622,
|
|
"mean_token_accuracy": 0.21056761592626572,
|
|
"num_tokens": 8516250.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.243447113037109,
|
|
"epoch": 4.108166576527853,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038358997292329406,
|
|
"loss": 4.8453,
|
|
"mean_token_accuracy": 0.21869916766881942,
|
|
"num_tokens": 8527220.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.288996839523316,
|
|
"epoch": 4.113574905354246,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00038321411636141214,
|
|
"loss": 4.9295,
|
|
"mean_token_accuracy": 0.2117052897810936,
|
|
"num_tokens": 8538037.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.279505634307862,
|
|
"epoch": 4.1189832341806385,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003828378665491811,
|
|
"loss": 4.8851,
|
|
"mean_token_accuracy": 0.2220398396253586,
|
|
"num_tokens": 8548684.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.201832723617554,
|
|
"epoch": 4.1243915630070305,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000382461224853894,
|
|
"loss": 4.7849,
|
|
"mean_token_accuracy": 0.21240828186273575,
|
|
"num_tokens": 8559756.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.252015161514282,
|
|
"epoch": 4.129799891833423,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003820841926442674,
|
|
"loss": 4.8359,
|
|
"mean_token_accuracy": 0.22234994322061538,
|
|
"num_tokens": 8570640.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.300533056259155,
|
|
"epoch": 4.135208220659816,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003817067712904372,
|
|
"loss": 4.8641,
|
|
"mean_token_accuracy": 0.2119019389152527,
|
|
"num_tokens": 8583127.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.218524503707886,
|
|
"epoch": 4.140616549486209,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038132896216395355,
|
|
"loss": 4.8431,
|
|
"mean_token_accuracy": 0.215255968272686,
|
|
"num_tokens": 8595213.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.2929524898529055,
|
|
"epoch": 4.146024878312601,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003809507666377758,
|
|
"loss": 4.8292,
|
|
"mean_token_accuracy": 0.22648167759180068,
|
|
"num_tokens": 8606057.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.317214155197144,
|
|
"epoch": 4.151433207138994,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038057218608626714,
|
|
"loss": 4.8973,
|
|
"mean_token_accuracy": 0.2176418885588646,
|
|
"num_tokens": 8617660.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.242032718658447,
|
|
"epoch": 4.156841535965387,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003801932218851903,
|
|
"loss": 4.8304,
|
|
"mean_token_accuracy": 0.21811017394065857,
|
|
"num_tokens": 8628595.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.300996351242065,
|
|
"epoch": 4.16224986479178,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000379813875411702,
|
|
"loss": 4.8862,
|
|
"mean_token_accuracy": 0.21567436754703523,
|
|
"num_tokens": 8639831.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.222472858428955,
|
|
"epoch": 4.167658193618172,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00037943414804434826,
|
|
"loss": 4.8264,
|
|
"mean_token_accuracy": 0.21448731273412705,
|
|
"num_tokens": 8650470.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.3071424007415775,
|
|
"epoch": 4.173066522444564,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003790540411630592,
|
|
"loss": 4.9432,
|
|
"mean_token_accuracy": 0.21564248353242874,
|
|
"num_tokens": 8660969.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.242237520217896,
|
|
"epoch": 4.178474851270957,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00037867355614914404,
|
|
"loss": 4.8173,
|
|
"mean_token_accuracy": 0.22410767823457717,
|
|
"num_tokens": 8670913.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.24915189743042,
|
|
"epoch": 4.18388318009735,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003782926943852864,
|
|
"loss": 4.9448,
|
|
"mean_token_accuracy": 0.20969100892543793,
|
|
"num_tokens": 8683028.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.383683586120606,
|
|
"epoch": 4.189291508923742,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00037791145725553853,
|
|
"loss": 5.0364,
|
|
"mean_token_accuracy": 0.20411381721496583,
|
|
"num_tokens": 8696454.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.346442365646363,
|
|
"epoch": 4.194699837750135,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003775298461453172,
|
|
"loss": 4.8215,
|
|
"mean_token_accuracy": 0.22135815173387527,
|
|
"num_tokens": 8707112.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.129479551315308,
|
|
"epoch": 4.200108166576528,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000377147862441398,
|
|
"loss": 4.8313,
|
|
"mean_token_accuracy": 0.22103142589330674,
|
|
"num_tokens": 8717696.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.271258401870727,
|
|
"epoch": 4.205516495402921,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00037676550753191075,
|
|
"loss": 4.7697,
|
|
"mean_token_accuracy": 0.22583625614643096,
|
|
"num_tokens": 8728571.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.28625979423523,
|
|
"epoch": 4.210924824229313,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003763827828063339,
|
|
"loss": 4.8873,
|
|
"mean_token_accuracy": 0.21592691242694856,
|
|
"num_tokens": 8739582.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.224824619293213,
|
|
"epoch": 4.2163331530557056,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00037599968965549,
|
|
"loss": 4.8511,
|
|
"mean_token_accuracy": 0.213607294857502,
|
|
"num_tokens": 8750527.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.2283659934997555,
|
|
"epoch": 4.221741481882098,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003756162294715405,
|
|
"loss": 4.8418,
|
|
"mean_token_accuracy": 0.2155377373099327,
|
|
"num_tokens": 8763140.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.24114761352539,
|
|
"epoch": 4.227149810708491,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00037523240364798063,
|
|
"loss": 4.868,
|
|
"mean_token_accuracy": 0.21395325064659118,
|
|
"num_tokens": 8774328.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.275230979919433,
|
|
"epoch": 4.232558139534884,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00037484821357963423,
|
|
"loss": 4.8906,
|
|
"mean_token_accuracy": 0.2126808375120163,
|
|
"num_tokens": 8785119.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.218007802963257,
|
|
"epoch": 4.237966468361276,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000374463660662649,
|
|
"loss": 4.8715,
|
|
"mean_token_accuracy": 0.21440057903528215,
|
|
"num_tokens": 8795833.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.254645538330078,
|
|
"epoch": 4.243374797187669,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000374078746294491,
|
|
"loss": 4.9608,
|
|
"mean_token_accuracy": 0.21045823395252228,
|
|
"num_tokens": 8807098.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.2683906078338625,
|
|
"epoch": 4.248783126014062,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00037369347187393996,
|
|
"loss": 4.77,
|
|
"mean_token_accuracy": 0.22449768483638763,
|
|
"num_tokens": 8818096.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.219247817993164,
|
|
"epoch": 4.254191454840455,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003733078388010842,
|
|
"loss": 4.9596,
|
|
"mean_token_accuracy": 0.19979992508888245,
|
|
"num_tokens": 8829451.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.319711828231812,
|
|
"epoch": 4.259599783666847,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003729218484773149,
|
|
"loss": 4.8784,
|
|
"mean_token_accuracy": 0.21760399043560028,
|
|
"num_tokens": 8840099.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.201181697845459,
|
|
"epoch": 4.2650081124932395,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003725355023053221,
|
|
"loss": 4.7973,
|
|
"mean_token_accuracy": 0.22008675336837769,
|
|
"num_tokens": 8851269.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.225205993652343,
|
|
"epoch": 4.270416441319632,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003721488016890884,
|
|
"loss": 4.7988,
|
|
"mean_token_accuracy": 0.22256304323673248,
|
|
"num_tokens": 8862871.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.280335855484009,
|
|
"epoch": 4.275824770146025,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00037176174803388496,
|
|
"loss": 4.8468,
|
|
"mean_token_accuracy": 0.21581949591636657,
|
|
"num_tokens": 8874328.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.243332195281982,
|
|
"epoch": 4.281233098972417,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00037137434274626545,
|
|
"loss": 4.853,
|
|
"mean_token_accuracy": 0.21557887494564057,
|
|
"num_tokens": 8885133.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.272100353240967,
|
|
"epoch": 4.28664142779881,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003709865872340618,
|
|
"loss": 4.8767,
|
|
"mean_token_accuracy": 0.21382254511117935,
|
|
"num_tokens": 8896479.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.235695219039917,
|
|
"epoch": 4.292049756625203,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003705984829063782,
|
|
"loss": 4.7784,
|
|
"mean_token_accuracy": 0.22039081454277037,
|
|
"num_tokens": 8907966.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.145072317123413,
|
|
"epoch": 4.297458085451596,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003702100311735869,
|
|
"loss": 4.8528,
|
|
"mean_token_accuracy": 0.20953898280858993,
|
|
"num_tokens": 8919042.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.267826795578003,
|
|
"epoch": 4.302866414277988,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00036982123344732223,
|
|
"loss": 4.8765,
|
|
"mean_token_accuracy": 0.2109612837433815,
|
|
"num_tokens": 8930456.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.220275831222534,
|
|
"epoch": 4.308274743104381,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00036943209114047613,
|
|
"loss": 4.937,
|
|
"mean_token_accuracy": 0.2152715101838112,
|
|
"num_tokens": 8941473.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.245625448226929,
|
|
"epoch": 4.3136830719307735,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003690426056671926,
|
|
"loss": 4.9444,
|
|
"mean_token_accuracy": 0.21144780218601228,
|
|
"num_tokens": 8952927.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.244691610336304,
|
|
"epoch": 4.319091400757166,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003686527784428627,
|
|
"loss": 4.9006,
|
|
"mean_token_accuracy": 0.21258871704339982,
|
|
"num_tokens": 8963831.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.294049453735352,
|
|
"epoch": 4.324499729583558,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00036826261088411955,
|
|
"loss": 4.8661,
|
|
"mean_token_accuracy": 0.2123163238167763,
|
|
"num_tokens": 8977147.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 4.324499729583558,
|
|
"eval_entropy": 5.227326540513472,
|
|
"eval_loss": 5.923820495605469,
|
|
"eval_mean_token_accuracy": 0.17602809323844584,
|
|
"eval_num_tokens": 8977147.0,
|
|
"eval_runtime": 2.5315,
|
|
"eval_samples_per_second": 1388.518,
|
|
"eval_steps_per_second": 173.812,
|
|
"step": 4000
|
|
}
|
|
],
|
|
"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": 1.4027172867072e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|