117c2fdc6a
Model: fpadovani/eng-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
6101 lines
165 KiB
JSON
6101 lines
165 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.243374797187669,
|
|
"eval_steps": 500,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 6.864051961898804,
|
|
"epoch": 0.005408328826392645,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 6.5257,
|
|
"mean_token_accuracy": 0.12957260981202126,
|
|
"num_tokens": 10942.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 6.855327892303467,
|
|
"epoch": 0.01081665765278529,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 6.4733,
|
|
"mean_token_accuracy": 0.1344677597284317,
|
|
"num_tokens": 21887.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 6.8918726444244385,
|
|
"epoch": 0.016224986479177934,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 6.6192,
|
|
"mean_token_accuracy": 0.12732454165816307,
|
|
"num_tokens": 34345.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 6.877776336669922,
|
|
"epoch": 0.02163331530557058,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 6.6279,
|
|
"mean_token_accuracy": 0.12559528797864913,
|
|
"num_tokens": 45445.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 6.873085641860962,
|
|
"epoch": 0.02704164413196322,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 6.6515,
|
|
"mean_token_accuracy": 0.12910161092877387,
|
|
"num_tokens": 58237.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 6.7455971240997314,
|
|
"epoch": 0.03244997295835587,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 6.421,
|
|
"mean_token_accuracy": 0.13589080199599265,
|
|
"num_tokens": 68098.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 6.782857465744018,
|
|
"epoch": 0.03785830178474851,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 6.5956,
|
|
"mean_token_accuracy": 0.1269644856452942,
|
|
"num_tokens": 79867.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 6.746934652328491,
|
|
"epoch": 0.04326663061114116,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 6.5788,
|
|
"mean_token_accuracy": 0.13147034570574762,
|
|
"num_tokens": 91257.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 6.78856520652771,
|
|
"epoch": 0.048674959437533805,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 6.6176,
|
|
"mean_token_accuracy": 0.11875083148479462,
|
|
"num_tokens": 103474.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 6.853682327270508,
|
|
"epoch": 0.05408328826392644,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 6.5603,
|
|
"mean_token_accuracy": 0.13152859807014466,
|
|
"num_tokens": 114830.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 6.866990423202514,
|
|
"epoch": 0.05949161709031909,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 6.5907,
|
|
"mean_token_accuracy": 0.1271680362522602,
|
|
"num_tokens": 126017.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 6.819481086730957,
|
|
"epoch": 0.06489994591671173,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 6.5497,
|
|
"mean_token_accuracy": 0.13208499029278756,
|
|
"num_tokens": 138509.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 6.824460887908936,
|
|
"epoch": 0.07030827474310439,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 6.6117,
|
|
"mean_token_accuracy": 0.12922831401228904,
|
|
"num_tokens": 148353.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 6.830938816070557,
|
|
"epoch": 0.07571660356949703,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 6.6626,
|
|
"mean_token_accuracy": 0.12804340198636055,
|
|
"num_tokens": 160269.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 6.792305183410645,
|
|
"epoch": 0.08112493239588967,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 6.4038,
|
|
"mean_token_accuracy": 0.12390550225973129,
|
|
"num_tokens": 171568.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 6.766619539260864,
|
|
"epoch": 0.08653326122228232,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 6.5291,
|
|
"mean_token_accuracy": 0.12462332919239998,
|
|
"num_tokens": 184243.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 6.779441785812378,
|
|
"epoch": 0.09194159004867496,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 6.5793,
|
|
"mean_token_accuracy": 0.12959954887628555,
|
|
"num_tokens": 194887.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 6.850970697402954,
|
|
"epoch": 0.09734991887506761,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 6.5453,
|
|
"mean_token_accuracy": 0.1317695237696171,
|
|
"num_tokens": 206476.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 6.879087781906128,
|
|
"epoch": 0.10275824770146025,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 6.5016,
|
|
"mean_token_accuracy": 0.13605281040072442,
|
|
"num_tokens": 216578.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 6.6328198432922365,
|
|
"epoch": 0.10816657652785289,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 6.4195,
|
|
"mean_token_accuracy": 0.12878495901823045,
|
|
"num_tokens": 226941.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 6.824092626571655,
|
|
"epoch": 0.11357490535424554,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 6.5031,
|
|
"mean_token_accuracy": 0.13698900192976,
|
|
"num_tokens": 238060.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 6.8112485885620115,
|
|
"epoch": 0.11898323418063818,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 6.4324,
|
|
"mean_token_accuracy": 0.1342177078127861,
|
|
"num_tokens": 249137.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 6.664848518371582,
|
|
"epoch": 0.12439156300703083,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 6.5648,
|
|
"mean_token_accuracy": 0.12625584304332732,
|
|
"num_tokens": 260712.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 6.752963876724243,
|
|
"epoch": 0.12979989183342347,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 6.3784,
|
|
"mean_token_accuracy": 0.13040537759661674,
|
|
"num_tokens": 271039.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 6.770984792709351,
|
|
"epoch": 0.1352082206598161,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 6.3524,
|
|
"mean_token_accuracy": 0.1345420315861702,
|
|
"num_tokens": 281353.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 6.684099960327148,
|
|
"epoch": 0.14061654948620878,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 6.4973,
|
|
"mean_token_accuracy": 0.12665031775832175,
|
|
"num_tokens": 292221.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 6.7672656059265135,
|
|
"epoch": 0.1460248783126014,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 6.4909,
|
|
"mean_token_accuracy": 0.12918338924646378,
|
|
"num_tokens": 303353.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 6.815447473526001,
|
|
"epoch": 0.15143320713899405,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 6.5593,
|
|
"mean_token_accuracy": 0.13174073919653892,
|
|
"num_tokens": 314165.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 6.7323511123657225,
|
|
"epoch": 0.1568415359653867,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 6.4191,
|
|
"mean_token_accuracy": 0.13412535786628724,
|
|
"num_tokens": 326387.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 6.795410251617431,
|
|
"epoch": 0.16224986479177933,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 6.5473,
|
|
"mean_token_accuracy": 0.12716327011585235,
|
|
"num_tokens": 339155.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 6.718407201766968,
|
|
"epoch": 0.167658193618172,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 6.4936,
|
|
"mean_token_accuracy": 0.13397022411227227,
|
|
"num_tokens": 350643.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 6.686786699295044,
|
|
"epoch": 0.17306652244456464,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 6.4321,
|
|
"mean_token_accuracy": 0.1322800897061825,
|
|
"num_tokens": 361977.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 6.820351600646973,
|
|
"epoch": 0.17847485127095727,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 6.5379,
|
|
"mean_token_accuracy": 0.1294139102101326,
|
|
"num_tokens": 373604.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 6.910545206069946,
|
|
"epoch": 0.1838831800973499,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 6.5585,
|
|
"mean_token_accuracy": 0.13030474483966828,
|
|
"num_tokens": 384306.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 6.647443962097168,
|
|
"epoch": 0.18929150892374255,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 6.4785,
|
|
"mean_token_accuracy": 0.1364760771393776,
|
|
"num_tokens": 396479.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 6.769079303741455,
|
|
"epoch": 0.19469983775013522,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 6.4207,
|
|
"mean_token_accuracy": 0.14159456118941308,
|
|
"num_tokens": 408489.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 6.721689462661743,
|
|
"epoch": 0.20010816657652786,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 6.4695,
|
|
"mean_token_accuracy": 0.13834813088178635,
|
|
"num_tokens": 419339.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 6.785417461395264,
|
|
"epoch": 0.2055164954029205,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 6.4534,
|
|
"mean_token_accuracy": 0.13452008813619615,
|
|
"num_tokens": 430486.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 6.692112255096435,
|
|
"epoch": 0.21092482422931313,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 6.5331,
|
|
"mean_token_accuracy": 0.12991808280348777,
|
|
"num_tokens": 440414.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 6.7671606063842775,
|
|
"epoch": 0.21633315305570577,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 6.5056,
|
|
"mean_token_accuracy": 0.12834976240992546,
|
|
"num_tokens": 451475.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 6.826629161834717,
|
|
"epoch": 0.22174148188209844,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 6.5449,
|
|
"mean_token_accuracy": 0.1341966614127159,
|
|
"num_tokens": 462053.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 6.835502099990845,
|
|
"epoch": 0.22714981070849108,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 6.5019,
|
|
"mean_token_accuracy": 0.13157635405659676,
|
|
"num_tokens": 473207.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 6.791409826278686,
|
|
"epoch": 0.23255813953488372,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.5007,
|
|
"mean_token_accuracy": 0.13225119560956955,
|
|
"num_tokens": 484999.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 6.8245649337768555,
|
|
"epoch": 0.23796646836127636,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 6.5292,
|
|
"mean_token_accuracy": 0.13114825561642646,
|
|
"num_tokens": 496772.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 6.761571311950684,
|
|
"epoch": 0.24337479718766902,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.5097,
|
|
"mean_token_accuracy": 0.1261039488017559,
|
|
"num_tokens": 507782.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 6.632718420028686,
|
|
"epoch": 0.24878312601406166,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.3935,
|
|
"mean_token_accuracy": 0.13307860344648362,
|
|
"num_tokens": 518950.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 6.848974227905273,
|
|
"epoch": 0.2541914548404543,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.4921,
|
|
"mean_token_accuracy": 0.13235593140125274,
|
|
"num_tokens": 530795.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 6.854920768737793,
|
|
"epoch": 0.25959978366684694,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.5274,
|
|
"mean_token_accuracy": 0.12910187914967536,
|
|
"num_tokens": 542878.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 6.648855829238892,
|
|
"epoch": 0.2650081124932396,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.4566,
|
|
"mean_token_accuracy": 0.13555580973625184,
|
|
"num_tokens": 553682.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 6.687378358840943,
|
|
"epoch": 0.2704164413196322,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.4573,
|
|
"mean_token_accuracy": 0.1364142492413521,
|
|
"num_tokens": 564433.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 6.809622764587402,
|
|
"epoch": 0.27582477014602486,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.5038,
|
|
"mean_token_accuracy": 0.1302446760237217,
|
|
"num_tokens": 576001.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 6.67371506690979,
|
|
"epoch": 0.28123309897241755,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.4197,
|
|
"mean_token_accuracy": 0.13783044442534448,
|
|
"num_tokens": 587138.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 6.641138887405395,
|
|
"epoch": 0.2866414277988102,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.4134,
|
|
"mean_token_accuracy": 0.13105393201112747,
|
|
"num_tokens": 597572.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 6.689500093460083,
|
|
"epoch": 0.2920497566252028,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.3849,
|
|
"mean_token_accuracy": 0.1401638939976692,
|
|
"num_tokens": 608879.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 6.763365650177002,
|
|
"epoch": 0.29745808545159547,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.3945,
|
|
"mean_token_accuracy": 0.1328580766916275,
|
|
"num_tokens": 620439.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 6.728887510299683,
|
|
"epoch": 0.3028664142779881,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.4887,
|
|
"mean_token_accuracy": 0.12868142277002334,
|
|
"num_tokens": 632543.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 6.729700183868408,
|
|
"epoch": 0.30827474310438074,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.5307,
|
|
"mean_token_accuracy": 0.1366984114050865,
|
|
"num_tokens": 643369.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 6.7040880680084225,
|
|
"epoch": 0.3136830719307734,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.4213,
|
|
"mean_token_accuracy": 0.13664653152227402,
|
|
"num_tokens": 655328.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 6.7930896282196045,
|
|
"epoch": 0.319091400757166,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.4706,
|
|
"mean_token_accuracy": 0.13196652084589006,
|
|
"num_tokens": 665780.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 6.654512739181518,
|
|
"epoch": 0.32449972958355866,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.3765,
|
|
"mean_token_accuracy": 0.1399880088865757,
|
|
"num_tokens": 676566.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 6.769630575180054,
|
|
"epoch": 0.3299080584099513,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.3971,
|
|
"mean_token_accuracy": 0.13649087622761727,
|
|
"num_tokens": 687407.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 6.615349292755127,
|
|
"epoch": 0.335316387236344,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.3185,
|
|
"mean_token_accuracy": 0.13677217811346054,
|
|
"num_tokens": 698162.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 6.621983480453491,
|
|
"epoch": 0.34072471606273663,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.295,
|
|
"mean_token_accuracy": 0.13893194645643234,
|
|
"num_tokens": 708743.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 6.840606784820556,
|
|
"epoch": 0.34613304488912927,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.5539,
|
|
"mean_token_accuracy": 0.13791529312729836,
|
|
"num_tokens": 719828.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 6.804727983474732,
|
|
"epoch": 0.3515413737155219,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.5244,
|
|
"mean_token_accuracy": 0.12768218517303467,
|
|
"num_tokens": 732514.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 6.586649608612061,
|
|
"epoch": 0.35694970254191455,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.3365,
|
|
"mean_token_accuracy": 0.13762425035238265,
|
|
"num_tokens": 742391.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 6.61699652671814,
|
|
"epoch": 0.3623580313683072,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.3468,
|
|
"mean_token_accuracy": 0.13812702521681786,
|
|
"num_tokens": 753638.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 6.8332000255584715,
|
|
"epoch": 0.3677663601946998,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.5721,
|
|
"mean_token_accuracy": 0.1307299703359604,
|
|
"num_tokens": 764850.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 6.6277220249176025,
|
|
"epoch": 0.37317468902109246,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.3783,
|
|
"mean_token_accuracy": 0.13669388741254807,
|
|
"num_tokens": 776192.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 6.656986141204834,
|
|
"epoch": 0.3785830178474851,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.3113,
|
|
"mean_token_accuracy": 0.1423935428261757,
|
|
"num_tokens": 787051.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 6.532338047027588,
|
|
"epoch": 0.3839913466738778,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.3685,
|
|
"mean_token_accuracy": 0.13058102130889893,
|
|
"num_tokens": 799281.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 6.699622631072998,
|
|
"epoch": 0.38939967550027044,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.387,
|
|
"mean_token_accuracy": 0.13420359566807746,
|
|
"num_tokens": 810003.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 6.604614973068237,
|
|
"epoch": 0.3948080043266631,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.3644,
|
|
"mean_token_accuracy": 0.13742000609636307,
|
|
"num_tokens": 819749.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 6.671362590789795,
|
|
"epoch": 0.4002163331530557,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.3468,
|
|
"mean_token_accuracy": 0.13851073011755943,
|
|
"num_tokens": 830552.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 6.74863977432251,
|
|
"epoch": 0.40562466197944835,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.4692,
|
|
"mean_token_accuracy": 0.13459497690200806,
|
|
"num_tokens": 840495.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 6.6008556365966795,
|
|
"epoch": 0.411032990805841,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.4267,
|
|
"mean_token_accuracy": 0.13428210094571114,
|
|
"num_tokens": 852915.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 6.708981895446778,
|
|
"epoch": 0.41644131963223363,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.3909,
|
|
"mean_token_accuracy": 0.13845089972019195,
|
|
"num_tokens": 862762.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 6.568316173553467,
|
|
"epoch": 0.42184964845862627,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.3296,
|
|
"mean_token_accuracy": 0.1392764002084732,
|
|
"num_tokens": 873023.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 6.658663272857666,
|
|
"epoch": 0.4272579772850189,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.4392,
|
|
"mean_token_accuracy": 0.1273169092833996,
|
|
"num_tokens": 885281.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 6.660024642944336,
|
|
"epoch": 0.43266630611141155,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.4295,
|
|
"mean_token_accuracy": 0.13264212384819984,
|
|
"num_tokens": 896210.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 6.573422765731811,
|
|
"epoch": 0.43807463493780424,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.3751,
|
|
"mean_token_accuracy": 0.13731420114636422,
|
|
"num_tokens": 909028.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 6.617922115325928,
|
|
"epoch": 0.4434829637641969,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.3025,
|
|
"mean_token_accuracy": 0.13610078319907187,
|
|
"num_tokens": 919710.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 6.681699609756469,
|
|
"epoch": 0.4488912925905895,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.3464,
|
|
"mean_token_accuracy": 0.13410636335611342,
|
|
"num_tokens": 930019.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 6.552076721191407,
|
|
"epoch": 0.45429962141698216,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.2992,
|
|
"mean_token_accuracy": 0.1357334055006504,
|
|
"num_tokens": 940535.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 6.447870922088623,
|
|
"epoch": 0.4597079502433748,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.2348,
|
|
"mean_token_accuracy": 0.13834500461816787,
|
|
"num_tokens": 951390.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 6.558719491958618,
|
|
"epoch": 0.46511627906976744,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.291,
|
|
"mean_token_accuracy": 0.13867247328162194,
|
|
"num_tokens": 963010.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 6.5435929775238035,
|
|
"epoch": 0.4705246078961601,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.2825,
|
|
"mean_token_accuracy": 0.14248405620455742,
|
|
"num_tokens": 974032.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.574478960037231,
|
|
"epoch": 0.4759329367225527,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.3202,
|
|
"mean_token_accuracy": 0.14242529571056367,
|
|
"num_tokens": 983954.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 6.665207958221435,
|
|
"epoch": 0.48134126554894535,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.3471,
|
|
"mean_token_accuracy": 0.14152060598134994,
|
|
"num_tokens": 994135.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.572685384750367,
|
|
"epoch": 0.48674959437533805,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.3247,
|
|
"mean_token_accuracy": 0.13299178332090378,
|
|
"num_tokens": 1006275.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.4127904891967775,
|
|
"epoch": 0.4921579232017307,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.1536,
|
|
"mean_token_accuracy": 0.14387866333127022,
|
|
"num_tokens": 1017767.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.568846130371094,
|
|
"epoch": 0.4975662520281233,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.2701,
|
|
"mean_token_accuracy": 0.14339046478271483,
|
|
"num_tokens": 1028286.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.528611373901367,
|
|
"epoch": 0.502974580854516,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.2962,
|
|
"mean_token_accuracy": 0.13511016070842743,
|
|
"num_tokens": 1040555.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.598082256317139,
|
|
"epoch": 0.5083829096809086,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.4092,
|
|
"mean_token_accuracy": 0.13908932209014893,
|
|
"num_tokens": 1053668.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.517001104354859,
|
|
"epoch": 0.5137912385073012,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.3556,
|
|
"mean_token_accuracy": 0.13394771441817283,
|
|
"num_tokens": 1066042.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.530339622497559,
|
|
"epoch": 0.5191995673336939,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.3735,
|
|
"mean_token_accuracy": 0.13361683785915374,
|
|
"num_tokens": 1077843.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.586634731292724,
|
|
"epoch": 0.5246078961600865,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.262,
|
|
"mean_token_accuracy": 0.13985889330506324,
|
|
"num_tokens": 1089448.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.548986101150513,
|
|
"epoch": 0.5300162249864792,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.3764,
|
|
"mean_token_accuracy": 0.1387718454003334,
|
|
"num_tokens": 1100989.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.645459794998169,
|
|
"epoch": 0.5354245538128718,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.3805,
|
|
"mean_token_accuracy": 0.13938464373350143,
|
|
"num_tokens": 1113012.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.53298978805542,
|
|
"epoch": 0.5408328826392644,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.3401,
|
|
"mean_token_accuracy": 0.13567214980721473,
|
|
"num_tokens": 1123551.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5408328826392644,
|
|
"eval_entropy": 6.77129298882051,
|
|
"eval_loss": 6.455276966094971,
|
|
"eval_mean_token_accuracy": 0.13739985009994019,
|
|
"eval_num_tokens": 1123551.0,
|
|
"eval_runtime": 2.5885,
|
|
"eval_samples_per_second": 1357.948,
|
|
"eval_steps_per_second": 169.985,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.616656875610351,
|
|
"epoch": 0.5462412114656571,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.2791,
|
|
"mean_token_accuracy": 0.14094609543681144,
|
|
"num_tokens": 1133838.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.4735087871551515,
|
|
"epoch": 0.5516495402920497,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.3233,
|
|
"mean_token_accuracy": 0.13018051609396936,
|
|
"num_tokens": 1146005.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.511937141418457,
|
|
"epoch": 0.5570578691184424,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.4183,
|
|
"mean_token_accuracy": 0.13036735653877257,
|
|
"num_tokens": 1157328.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.5359978675842285,
|
|
"epoch": 0.5624661979448351,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.2176,
|
|
"mean_token_accuracy": 0.1424185998737812,
|
|
"num_tokens": 1168592.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.532036399841308,
|
|
"epoch": 0.5678745267712277,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.3445,
|
|
"mean_token_accuracy": 0.13651241958141327,
|
|
"num_tokens": 1178680.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.533176612854004,
|
|
"epoch": 0.5732828555976204,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.3,
|
|
"mean_token_accuracy": 0.1420231744647026,
|
|
"num_tokens": 1188825.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.53720064163208,
|
|
"epoch": 0.578691184424013,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.3448,
|
|
"mean_token_accuracy": 0.13785845115780831,
|
|
"num_tokens": 1201375.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.602963924407959,
|
|
"epoch": 0.5840995132504057,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.4732,
|
|
"mean_token_accuracy": 0.13540672287344932,
|
|
"num_tokens": 1211915.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.56973090171814,
|
|
"epoch": 0.5895078420767983,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.3834,
|
|
"mean_token_accuracy": 0.14111291915178298,
|
|
"num_tokens": 1223151.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.601230239868164,
|
|
"epoch": 0.5949161709031909,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.2881,
|
|
"mean_token_accuracy": 0.14251122027635574,
|
|
"num_tokens": 1234229.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.462851238250733,
|
|
"epoch": 0.6003244997295836,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.2863,
|
|
"mean_token_accuracy": 0.14203285053372383,
|
|
"num_tokens": 1245298.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.57589693069458,
|
|
"epoch": 0.6057328285559762,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.3156,
|
|
"mean_token_accuracy": 0.1349334165453911,
|
|
"num_tokens": 1256154.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.442286968231201,
|
|
"epoch": 0.6111411573823688,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.2135,
|
|
"mean_token_accuracy": 0.14049630761146545,
|
|
"num_tokens": 1267404.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.529563570022583,
|
|
"epoch": 0.6165494862087615,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.2749,
|
|
"mean_token_accuracy": 0.13832947835326195,
|
|
"num_tokens": 1278894.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.463529872894287,
|
|
"epoch": 0.6219578150351541,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.2662,
|
|
"mean_token_accuracy": 0.14242759048938752,
|
|
"num_tokens": 1290408.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.492969179153443,
|
|
"epoch": 0.6273661438615468,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.3411,
|
|
"mean_token_accuracy": 0.13858584463596343,
|
|
"num_tokens": 1301508.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.571087789535523,
|
|
"epoch": 0.6327744726879394,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.3059,
|
|
"mean_token_accuracy": 0.1433561213314533,
|
|
"num_tokens": 1312551.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.517476177215576,
|
|
"epoch": 0.638182801514332,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.4164,
|
|
"mean_token_accuracy": 0.1359536699950695,
|
|
"num_tokens": 1324876.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.597873401641846,
|
|
"epoch": 0.6435911303407247,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.3011,
|
|
"mean_token_accuracy": 0.14250493571162223,
|
|
"num_tokens": 1335401.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.4096986770629885,
|
|
"epoch": 0.6489994591671173,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.2329,
|
|
"mean_token_accuracy": 0.13733577728271484,
|
|
"num_tokens": 1348265.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.3921589851379395,
|
|
"epoch": 0.65440778799351,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.1839,
|
|
"mean_token_accuracy": 0.1522812619805336,
|
|
"num_tokens": 1359078.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.281019926071167,
|
|
"epoch": 0.6598161168199026,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.1089,
|
|
"mean_token_accuracy": 0.1434238463640213,
|
|
"num_tokens": 1369499.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.404446697235107,
|
|
"epoch": 0.6652244456462953,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.206,
|
|
"mean_token_accuracy": 0.13861049264669417,
|
|
"num_tokens": 1380329.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.478784847259521,
|
|
"epoch": 0.670632774472688,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.3124,
|
|
"mean_token_accuracy": 0.13905679434537888,
|
|
"num_tokens": 1393047.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.448394870758056,
|
|
"epoch": 0.6760411032990806,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.2976,
|
|
"mean_token_accuracy": 0.13759241476655007,
|
|
"num_tokens": 1405600.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.438097906112671,
|
|
"epoch": 0.6814494321254733,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.2289,
|
|
"mean_token_accuracy": 0.14040346965193748,
|
|
"num_tokens": 1416670.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.408063554763794,
|
|
"epoch": 0.6868577609518659,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.254,
|
|
"mean_token_accuracy": 0.14151675775647163,
|
|
"num_tokens": 1428929.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.448219919204712,
|
|
"epoch": 0.6922660897782585,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.1646,
|
|
"mean_token_accuracy": 0.1440845735371113,
|
|
"num_tokens": 1439298.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.489011144638061,
|
|
"epoch": 0.6976744186046512,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.257,
|
|
"mean_token_accuracy": 0.14002981781959534,
|
|
"num_tokens": 1449853.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.411071252822876,
|
|
"epoch": 0.7030827474310438,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.3471,
|
|
"mean_token_accuracy": 0.1383639879524708,
|
|
"num_tokens": 1460816.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.382777166366577,
|
|
"epoch": 0.7084910762574365,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.2017,
|
|
"mean_token_accuracy": 0.13986401185393332,
|
|
"num_tokens": 1473003.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.459758758544922,
|
|
"epoch": 0.7138994050838291,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.3254,
|
|
"mean_token_accuracy": 0.13349702432751656,
|
|
"num_tokens": 1484745.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.442171239852906,
|
|
"epoch": 0.7193077339102217,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.2002,
|
|
"mean_token_accuracy": 0.1414022244513035,
|
|
"num_tokens": 1495803.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.398311376571655,
|
|
"epoch": 0.7247160627366144,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.1889,
|
|
"mean_token_accuracy": 0.14995323568582536,
|
|
"num_tokens": 1507504.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.462614250183106,
|
|
"epoch": 0.730124391563007,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.3008,
|
|
"mean_token_accuracy": 0.13920977488160133,
|
|
"num_tokens": 1518765.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.307487154006958,
|
|
"epoch": 0.7355327203893997,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.1895,
|
|
"mean_token_accuracy": 0.14220963269472123,
|
|
"num_tokens": 1529481.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.427918529510498,
|
|
"epoch": 0.7409410492157923,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.2088,
|
|
"mean_token_accuracy": 0.1430448904633522,
|
|
"num_tokens": 1539612.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.441134834289551,
|
|
"epoch": 0.7463493780421849,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.3807,
|
|
"mean_token_accuracy": 0.13819732442498206,
|
|
"num_tokens": 1551370.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.452056884765625,
|
|
"epoch": 0.7517577068685776,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.2674,
|
|
"mean_token_accuracy": 0.14520786255598067,
|
|
"num_tokens": 1561662.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.388766479492188,
|
|
"epoch": 0.7571660356949702,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.2264,
|
|
"mean_token_accuracy": 0.1406054303050041,
|
|
"num_tokens": 1573479.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.406115865707397,
|
|
"epoch": 0.7625743645213628,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.1794,
|
|
"mean_token_accuracy": 0.1461093693971634,
|
|
"num_tokens": 1583684.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.343323850631714,
|
|
"epoch": 0.7679826933477556,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.2413,
|
|
"mean_token_accuracy": 0.1426021210849285,
|
|
"num_tokens": 1595694.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.319337940216064,
|
|
"epoch": 0.7733910221741482,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.2267,
|
|
"mean_token_accuracy": 0.14130815863609314,
|
|
"num_tokens": 1606104.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.439924716949463,
|
|
"epoch": 0.7787993510005409,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.3083,
|
|
"mean_token_accuracy": 0.141408871114254,
|
|
"num_tokens": 1617368.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.47912859916687,
|
|
"epoch": 0.7842076798269335,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.3055,
|
|
"mean_token_accuracy": 0.13773123025894166,
|
|
"num_tokens": 1630485.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.369973373413086,
|
|
"epoch": 0.7896160086533262,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.1384,
|
|
"mean_token_accuracy": 0.14500918239355087,
|
|
"num_tokens": 1641411.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.363615798950195,
|
|
"epoch": 0.7950243374797188,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.1572,
|
|
"mean_token_accuracy": 0.1470560222864151,
|
|
"num_tokens": 1653737.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.4537757396697994,
|
|
"epoch": 0.8004326663061114,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.3582,
|
|
"mean_token_accuracy": 0.13919734507799147,
|
|
"num_tokens": 1665910.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.295696640014649,
|
|
"epoch": 0.8058409951325041,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.1248,
|
|
"mean_token_accuracy": 0.14440400525927544,
|
|
"num_tokens": 1676927.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.3269532203674315,
|
|
"epoch": 0.8112493239588967,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.1839,
|
|
"mean_token_accuracy": 0.13821593970060347,
|
|
"num_tokens": 1688167.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.300651693344117,
|
|
"epoch": 0.8166576527852893,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.2722,
|
|
"mean_token_accuracy": 0.1447138175368309,
|
|
"num_tokens": 1699899.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.397444534301758,
|
|
"epoch": 0.822065981611682,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.188,
|
|
"mean_token_accuracy": 0.1506570905447006,
|
|
"num_tokens": 1709862.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.273593282699585,
|
|
"epoch": 0.8274743104380746,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.1913,
|
|
"mean_token_accuracy": 0.14058721661567689,
|
|
"num_tokens": 1719902.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.319923543930054,
|
|
"epoch": 0.8328826392644673,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.1877,
|
|
"mean_token_accuracy": 0.14031919091939926,
|
|
"num_tokens": 1730351.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.278684139251709,
|
|
"epoch": 0.8382909680908599,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.1771,
|
|
"mean_token_accuracy": 0.13964096456766129,
|
|
"num_tokens": 1741783.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.278208160400391,
|
|
"epoch": 0.8436992969172525,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.2118,
|
|
"mean_token_accuracy": 0.14272338300943374,
|
|
"num_tokens": 1752952.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.494202136993408,
|
|
"epoch": 0.8491076257436452,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.2294,
|
|
"mean_token_accuracy": 0.14174022600054742,
|
|
"num_tokens": 1764014.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.284159898757935,
|
|
"epoch": 0.8545159545700378,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.1819,
|
|
"mean_token_accuracy": 0.1421780988574028,
|
|
"num_tokens": 1776154.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.297621440887451,
|
|
"epoch": 0.8599242833964305,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.2507,
|
|
"mean_token_accuracy": 0.13820455372333526,
|
|
"num_tokens": 1787292.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.3149923324584964,
|
|
"epoch": 0.8653326122228231,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.223,
|
|
"mean_token_accuracy": 0.14442499577999116,
|
|
"num_tokens": 1799469.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.235135173797607,
|
|
"epoch": 0.8707409410492158,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.0925,
|
|
"mean_token_accuracy": 0.15308079421520232,
|
|
"num_tokens": 1810091.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.37100772857666,
|
|
"epoch": 0.8761492698756085,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.2464,
|
|
"mean_token_accuracy": 0.14540659710764886,
|
|
"num_tokens": 1820835.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.300705146789551,
|
|
"epoch": 0.8815575987020011,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.2155,
|
|
"mean_token_accuracy": 0.14222077131271363,
|
|
"num_tokens": 1831868.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.262601566314697,
|
|
"epoch": 0.8869659275283938,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.1819,
|
|
"mean_token_accuracy": 0.14557501971721648,
|
|
"num_tokens": 1842040.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.428664112091065,
|
|
"epoch": 0.8923742563547864,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.2513,
|
|
"mean_token_accuracy": 0.14266791865229606,
|
|
"num_tokens": 1853765.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.31456356048584,
|
|
"epoch": 0.897782585181179,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.2832,
|
|
"mean_token_accuracy": 0.14433787241578103,
|
|
"num_tokens": 1866081.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.301344776153565,
|
|
"epoch": 0.9031909140075717,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.1803,
|
|
"mean_token_accuracy": 0.14564810693264008,
|
|
"num_tokens": 1876754.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.383733701705933,
|
|
"epoch": 0.9085992428339643,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.2967,
|
|
"mean_token_accuracy": 0.1391839236021042,
|
|
"num_tokens": 1888299.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.140969562530517,
|
|
"epoch": 0.914007571660357,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.0718,
|
|
"mean_token_accuracy": 0.14790207147598267,
|
|
"num_tokens": 1899027.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.312633848190307,
|
|
"epoch": 0.9194159004867496,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.1972,
|
|
"mean_token_accuracy": 0.14347611665725707,
|
|
"num_tokens": 1911043.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.2559020042419435,
|
|
"epoch": 0.9248242293131422,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.127,
|
|
"mean_token_accuracy": 0.14566482454538346,
|
|
"num_tokens": 1921295.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.304692983627319,
|
|
"epoch": 0.9302325581395349,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.2181,
|
|
"mean_token_accuracy": 0.1475878342986107,
|
|
"num_tokens": 1932944.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.239846420288086,
|
|
"epoch": 0.9356408869659275,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.1703,
|
|
"mean_token_accuracy": 0.13581379726529122,
|
|
"num_tokens": 1944072.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.267998695373535,
|
|
"epoch": 0.9410492157923201,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.1739,
|
|
"mean_token_accuracy": 0.15081078559160233,
|
|
"num_tokens": 1954798.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.269789123535157,
|
|
"epoch": 0.9464575446187128,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.119,
|
|
"mean_token_accuracy": 0.15551341474056243,
|
|
"num_tokens": 1965304.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.316447162628174,
|
|
"epoch": 0.9518658734451054,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.185,
|
|
"mean_token_accuracy": 0.14262910634279252,
|
|
"num_tokens": 1976797.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.195334625244141,
|
|
"epoch": 0.9572742022714981,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.2027,
|
|
"mean_token_accuracy": 0.1425718829035759,
|
|
"num_tokens": 1987448.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.126097774505615,
|
|
"epoch": 0.9626825310978907,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.0369,
|
|
"mean_token_accuracy": 0.13968317657709123,
|
|
"num_tokens": 2000948.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.29197359085083,
|
|
"epoch": 0.9680908599242833,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.1307,
|
|
"mean_token_accuracy": 0.14575395435094834,
|
|
"num_tokens": 2011807.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.23886775970459,
|
|
"epoch": 0.9734991887506761,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.2025,
|
|
"mean_token_accuracy": 0.1369840271770954,
|
|
"num_tokens": 2022835.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.1420495986938475,
|
|
"epoch": 0.9789075175770687,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.1175,
|
|
"mean_token_accuracy": 0.1438199281692505,
|
|
"num_tokens": 2034331.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.316672372817993,
|
|
"epoch": 0.9843158464034614,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.1172,
|
|
"mean_token_accuracy": 0.1416486456990242,
|
|
"num_tokens": 2047002.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.058028793334961,
|
|
"epoch": 0.989724175229854,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.0726,
|
|
"mean_token_accuracy": 0.14620630741119384,
|
|
"num_tokens": 2056299.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.255609464645386,
|
|
"epoch": 0.9951325040562466,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.1849,
|
|
"mean_token_accuracy": 0.1430879294872284,
|
|
"num_tokens": 2068031.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.221119668748644,
|
|
"epoch": 1.0,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.1729,
|
|
"mean_token_accuracy": 0.14627348466051948,
|
|
"num_tokens": 2076889.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.230249977111816,
|
|
"epoch": 1.0054083288263926,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 5.9619,
|
|
"mean_token_accuracy": 0.1546672150492668,
|
|
"num_tokens": 2087316.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.195207643508911,
|
|
"epoch": 1.0108166576527853,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 5.9523,
|
|
"mean_token_accuracy": 0.14858346730470656,
|
|
"num_tokens": 2098522.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.0610815525054935,
|
|
"epoch": 1.016224986479178,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 5.9436,
|
|
"mean_token_accuracy": 0.1535676211118698,
|
|
"num_tokens": 2110015.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.187965297698975,
|
|
"epoch": 1.0216333153055706,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.0458,
|
|
"mean_token_accuracy": 0.14821772575378417,
|
|
"num_tokens": 2121333.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.1689681053161625,
|
|
"epoch": 1.0270416441319632,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 5.9325,
|
|
"mean_token_accuracy": 0.15763497352600098,
|
|
"num_tokens": 2131385.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.1999763488769535,
|
|
"epoch": 1.0324499729583558,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 5.9699,
|
|
"mean_token_accuracy": 0.15700915455818176,
|
|
"num_tokens": 2143274.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.015431594848633,
|
|
"epoch": 1.0378583017847485,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 5.875,
|
|
"mean_token_accuracy": 0.15367670357227325,
|
|
"num_tokens": 2154529.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.182692766189575,
|
|
"epoch": 1.043266630611141,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.0147,
|
|
"mean_token_accuracy": 0.15401246696710585,
|
|
"num_tokens": 2165949.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.116761016845703,
|
|
"epoch": 1.0486749594375337,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.008,
|
|
"mean_token_accuracy": 0.15102231055498122,
|
|
"num_tokens": 2176710.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.13912878036499,
|
|
"epoch": 1.0540832882639264,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 5.9699,
|
|
"mean_token_accuracy": 0.15905411243438722,
|
|
"num_tokens": 2189470.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.046469640731812,
|
|
"epoch": 1.059491617090319,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 5.9482,
|
|
"mean_token_accuracy": 0.1573230430483818,
|
|
"num_tokens": 2200266.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.112458515167236,
|
|
"epoch": 1.0648999459167117,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 5.9546,
|
|
"mean_token_accuracy": 0.15503815859556197,
|
|
"num_tokens": 2210736.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.165264558792114,
|
|
"epoch": 1.0703082747431043,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.0686,
|
|
"mean_token_accuracy": 0.14824794977903366,
|
|
"num_tokens": 2221057.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.161154079437256,
|
|
"epoch": 1.075716603569497,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.0536,
|
|
"mean_token_accuracy": 0.14784543812274933,
|
|
"num_tokens": 2232352.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.177089881896973,
|
|
"epoch": 1.0811249323958896,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 5.9584,
|
|
"mean_token_accuracy": 0.1536453515291214,
|
|
"num_tokens": 2244655.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0811249323958896,
|
|
"eval_entropy": 5.870350288261067,
|
|
"eval_loss": 6.289115905761719,
|
|
"eval_mean_token_accuracy": 0.14856769407845355,
|
|
"eval_num_tokens": 2244655.0,
|
|
"eval_runtime": 2.521,
|
|
"eval_samples_per_second": 1394.308,
|
|
"eval_steps_per_second": 174.536,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.018855476379395,
|
|
"epoch": 1.0865332612222822,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999997383518051,
|
|
"loss": 5.9204,
|
|
"mean_token_accuracy": 0.15997352302074433,
|
|
"num_tokens": 2255699.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.0613984107971195,
|
|
"epoch": 1.0919415900486749,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999986754070562,
|
|
"loss": 5.8996,
|
|
"mean_token_accuracy": 0.15893014371395112,
|
|
"num_tokens": 2266627.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.199114370346069,
|
|
"epoch": 1.0973499188750675,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004999967948166009,
|
|
"loss": 6.085,
|
|
"mean_token_accuracy": 0.1514703169465065,
|
|
"num_tokens": 2277814.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.16807861328125,
|
|
"epoch": 1.1027582477014604,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999940965872734,
|
|
"loss": 6.0391,
|
|
"mean_token_accuracy": 0.15074968487024307,
|
|
"num_tokens": 2287344.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 5.929051113128662,
|
|
"epoch": 1.1081665765278528,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999905807288788,
|
|
"loss": 5.8149,
|
|
"mean_token_accuracy": 0.16056424379348755,
|
|
"num_tokens": 2298332.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.119269990921021,
|
|
"epoch": 1.1135749053542456,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499986247254194,
|
|
"loss": 5.9508,
|
|
"mean_token_accuracy": 0.15462301671504974,
|
|
"num_tokens": 2309161.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.005192136764526,
|
|
"epoch": 1.1189832341806383,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004999810961789667,
|
|
"loss": 5.9673,
|
|
"mean_token_accuracy": 0.1558970808982849,
|
|
"num_tokens": 2319516.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.02630295753479,
|
|
"epoch": 1.124391563007031,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499975127521916,
|
|
"loss": 5.8491,
|
|
"mean_token_accuracy": 0.16057343631982804,
|
|
"num_tokens": 2330441.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.018536949157715,
|
|
"epoch": 1.1297998918334236,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999683413047319,
|
|
"loss": 5.8623,
|
|
"mean_token_accuracy": 0.16048752069473265,
|
|
"num_tokens": 2340524.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.028919363021851,
|
|
"epoch": 1.1352082206598162,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999607375520759,
|
|
"loss": 5.8661,
|
|
"mean_token_accuracy": 0.15480372905731202,
|
|
"num_tokens": 2353379.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.055306529998779,
|
|
"epoch": 1.1406165494862088,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999523162915794,
|
|
"loss": 6.0438,
|
|
"mean_token_accuracy": 0.1485805407166481,
|
|
"num_tokens": 2365708.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.046649026870727,
|
|
"epoch": 1.1460248783126015,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999430775538459,
|
|
"loss": 5.9392,
|
|
"mean_token_accuracy": 0.1528744824230671,
|
|
"num_tokens": 2375630.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.007912063598633,
|
|
"epoch": 1.151433207138994,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999330213724487,
|
|
"loss": 5.952,
|
|
"mean_token_accuracy": 0.15200790762901306,
|
|
"num_tokens": 2386885.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.064875078201294,
|
|
"epoch": 1.1568415359653867,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999221477839319,
|
|
"loss": 5.9886,
|
|
"mean_token_accuracy": 0.15056426972150802,
|
|
"num_tokens": 2397676.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.106745052337646,
|
|
"epoch": 1.1622498647917794,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999104568278104,
|
|
"loss": 5.9936,
|
|
"mean_token_accuracy": 0.1488187149167061,
|
|
"num_tokens": 2409039.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.072894811630249,
|
|
"epoch": 1.167658193618172,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000499897948546569,
|
|
"loss": 6.0203,
|
|
"mean_token_accuracy": 0.1492675483226776,
|
|
"num_tokens": 2420103.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 5.997014904022217,
|
|
"epoch": 1.1730665224445647,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004998846229856629,
|
|
"loss": 5.9508,
|
|
"mean_token_accuracy": 0.1516289785504341,
|
|
"num_tokens": 2430792.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.076592969894409,
|
|
"epoch": 1.1784748512709573,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998704801935171,
|
|
"loss": 5.9402,
|
|
"mean_token_accuracy": 0.15488605797290803,
|
|
"num_tokens": 2442165.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.0133076190948485,
|
|
"epoch": 1.18388318009735,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998555202215267,
|
|
"loss": 5.8876,
|
|
"mean_token_accuracy": 0.1536034032702446,
|
|
"num_tokens": 2454231.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.003773021697998,
|
|
"epoch": 1.1892915089237426,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998397431240561,
|
|
"loss": 5.9157,
|
|
"mean_token_accuracy": 0.1577537938952446,
|
|
"num_tokens": 2464828.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.016058254241943,
|
|
"epoch": 1.1946998377501352,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998231489584395,
|
|
"loss": 5.9751,
|
|
"mean_token_accuracy": 0.1452740930020809,
|
|
"num_tokens": 2476144.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.176415681838989,
|
|
"epoch": 1.2001081665765279,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00049980573778498,
|
|
"loss": 6.152,
|
|
"mean_token_accuracy": 0.14584484994411467,
|
|
"num_tokens": 2489116.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.145106458663941,
|
|
"epoch": 1.2055164954029205,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004997875096669498,
|
|
"loss": 6.0119,
|
|
"mean_token_accuracy": 0.15233825743198395,
|
|
"num_tokens": 2501609.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 5.896210384368897,
|
|
"epoch": 1.2109248242293131,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004997684646705901,
|
|
"loss": 5.9242,
|
|
"mean_token_accuracy": 0.16018943190574647,
|
|
"num_tokens": 2513338.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.141838550567627,
|
|
"epoch": 1.2163331530557058,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004997486028651103,
|
|
"loss": 5.9287,
|
|
"mean_token_accuracy": 0.16149964630603791,
|
|
"num_tokens": 2523199.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.035762691497803,
|
|
"epoch": 1.2217414818820984,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997279243226884,
|
|
"loss": 5.9464,
|
|
"mean_token_accuracy": 0.1545608699321747,
|
|
"num_tokens": 2535136.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 5.964288520812988,
|
|
"epoch": 1.227149810708491,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997064291184701,
|
|
"loss": 5.8682,
|
|
"mean_token_accuracy": 0.15200766772031785,
|
|
"num_tokens": 2546428.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.1283605098724365,
|
|
"epoch": 1.2325581395348837,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004996841173305691,
|
|
"loss": 6.0492,
|
|
"mean_token_accuracy": 0.1529114067554474,
|
|
"num_tokens": 2558304.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 5.852491044998169,
|
|
"epoch": 1.2379664683612763,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004996609890400665,
|
|
"loss": 5.7463,
|
|
"mean_token_accuracy": 0.1669026330113411,
|
|
"num_tokens": 2568717.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.0190393924713135,
|
|
"epoch": 1.243374797187669,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004996370443310107,
|
|
"loss": 5.9566,
|
|
"mean_token_accuracy": 0.14948671758174897,
|
|
"num_tokens": 2578748.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.08936882019043,
|
|
"epoch": 1.2487831260140616,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004996122832904165,
|
|
"loss": 5.919,
|
|
"mean_token_accuracy": 0.15417942851781846,
|
|
"num_tokens": 2590095.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 5.994193267822266,
|
|
"epoch": 1.2541914548404542,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004995867060082658,
|
|
"loss": 5.938,
|
|
"mean_token_accuracy": 0.15805463641881942,
|
|
"num_tokens": 2600929.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.03876519203186,
|
|
"epoch": 1.2595997836668469,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004995603125775065,
|
|
"loss": 5.9526,
|
|
"mean_token_accuracy": 0.1548033908009529,
|
|
"num_tokens": 2611830.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 5.966577386856079,
|
|
"epoch": 1.2650081124932395,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004995331030940523,
|
|
"loss": 5.9084,
|
|
"mean_token_accuracy": 0.15348549783229828,
|
|
"num_tokens": 2623177.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 5.9641149044036865,
|
|
"epoch": 1.2704164413196322,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004995050776567826,
|
|
"loss": 5.9033,
|
|
"mean_token_accuracy": 0.15433114916086196,
|
|
"num_tokens": 2635563.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.030624532699585,
|
|
"epoch": 1.2758247701460248,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004994762363675421,
|
|
"loss": 5.8889,
|
|
"mean_token_accuracy": 0.16223973035812378,
|
|
"num_tokens": 2647157.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 5.937535953521729,
|
|
"epoch": 1.2812330989724177,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004994465793311396,
|
|
"loss": 5.8659,
|
|
"mean_token_accuracy": 0.1634291410446167,
|
|
"num_tokens": 2658175.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 5.9314147472381595,
|
|
"epoch": 1.28664142779881,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004994161066553493,
|
|
"loss": 5.8247,
|
|
"mean_token_accuracy": 0.16096442192792892,
|
|
"num_tokens": 2669458.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.002810955047607,
|
|
"epoch": 1.292049756625203,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004993848184509086,
|
|
"loss": 6.0349,
|
|
"mean_token_accuracy": 0.15508441999554634,
|
|
"num_tokens": 2679727.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.0271283149719235,
|
|
"epoch": 1.2974580854515954,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004993527148315193,
|
|
"loss": 5.9989,
|
|
"mean_token_accuracy": 0.15247225612401963,
|
|
"num_tokens": 2691031.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 5.975004196166992,
|
|
"epoch": 1.3028664142779882,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004993197959138458,
|
|
"loss": 5.7931,
|
|
"mean_token_accuracy": 0.1641109436750412,
|
|
"num_tokens": 2701699.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 5.905522966384888,
|
|
"epoch": 1.3082747431043806,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004992860618175156,
|
|
"loss": 5.8418,
|
|
"mean_token_accuracy": 0.1618697851896286,
|
|
"num_tokens": 2713038.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 5.8916487216949465,
|
|
"epoch": 1.3136830719307735,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004992515126651184,
|
|
"loss": 5.8244,
|
|
"mean_token_accuracy": 0.162429741024971,
|
|
"num_tokens": 2723685.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 5.947316837310791,
|
|
"epoch": 1.319091400757166,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004992161485822059,
|
|
"loss": 5.9178,
|
|
"mean_token_accuracy": 0.15724743157625198,
|
|
"num_tokens": 2735341.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 5.998585796356201,
|
|
"epoch": 1.3244997295835588,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004991799696972913,
|
|
"loss": 5.8612,
|
|
"mean_token_accuracy": 0.16257011592388154,
|
|
"num_tokens": 2746052.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 5.895526170730591,
|
|
"epoch": 1.3299080584099512,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004991429761418488,
|
|
"loss": 5.8371,
|
|
"mean_token_accuracy": 0.16027999371290208,
|
|
"num_tokens": 2757155.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 5.961960029602051,
|
|
"epoch": 1.335316387236344,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499105168050313,
|
|
"loss": 5.9027,
|
|
"mean_token_accuracy": 0.15204052925109862,
|
|
"num_tokens": 2769631.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.098041534423828,
|
|
"epoch": 1.3407247160627367,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004990665455600787,
|
|
"loss": 6.0006,
|
|
"mean_token_accuracy": 0.15507472008466722,
|
|
"num_tokens": 2780797.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 5.862522077560425,
|
|
"epoch": 1.3461330448891293,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004990271088114999,
|
|
"loss": 5.8481,
|
|
"mean_token_accuracy": 0.15496804267168046,
|
|
"num_tokens": 2792721.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.012833166122436,
|
|
"epoch": 1.351541373715522,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004989868579478901,
|
|
"loss": 5.9086,
|
|
"mean_token_accuracy": 0.16479166895151137,
|
|
"num_tokens": 2804531.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 5.9520423412323,
|
|
"epoch": 1.3569497025419146,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004989457931155209,
|
|
"loss": 5.9031,
|
|
"mean_token_accuracy": 0.15717396438121795,
|
|
"num_tokens": 2816666.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 5.988644409179687,
|
|
"epoch": 1.3623580313683072,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004989039144636219,
|
|
"loss": 5.933,
|
|
"mean_token_accuracy": 0.16185724437236787,
|
|
"num_tokens": 2827977.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 5.929494953155517,
|
|
"epoch": 1.3677663601946999,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004988612221443804,
|
|
"loss": 5.9101,
|
|
"mean_token_accuracy": 0.15438521206378936,
|
|
"num_tokens": 2839069.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.02064790725708,
|
|
"epoch": 1.3731746890210925,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004988177163129403,
|
|
"loss": 5.9946,
|
|
"mean_token_accuracy": 0.14949484094977378,
|
|
"num_tokens": 2850632.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 5.903779649734497,
|
|
"epoch": 1.3785830178474852,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004987733971274018,
|
|
"loss": 5.8714,
|
|
"mean_token_accuracy": 0.15512411892414094,
|
|
"num_tokens": 2862204.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 5.927542591094971,
|
|
"epoch": 1.3839913466738778,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004987282647488211,
|
|
"loss": 5.8327,
|
|
"mean_token_accuracy": 0.16609278023242952,
|
|
"num_tokens": 2874010.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 5.962044954299927,
|
|
"epoch": 1.3893996755002704,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004986823193412093,
|
|
"loss": 5.8952,
|
|
"mean_token_accuracy": 0.16253606528043746,
|
|
"num_tokens": 2884315.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.061125898361206,
|
|
"epoch": 1.394808004326663,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000498635561071532,
|
|
"loss": 6.0569,
|
|
"mean_token_accuracy": 0.15368083566427232,
|
|
"num_tokens": 2896327.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 5.895346450805664,
|
|
"epoch": 1.4002163331530557,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000498587990109709,
|
|
"loss": 5.8406,
|
|
"mean_token_accuracy": 0.15671723783016206,
|
|
"num_tokens": 2907007.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 5.933822584152222,
|
|
"epoch": 1.4056246619794484,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004985396066286133,
|
|
"loss": 5.8617,
|
|
"mean_token_accuracy": 0.16459044367074965,
|
|
"num_tokens": 2918852.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 5.91065239906311,
|
|
"epoch": 1.411032990805841,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004984904108040706,
|
|
"loss": 5.786,
|
|
"mean_token_accuracy": 0.15888032019138337,
|
|
"num_tokens": 2929948.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.868727111816407,
|
|
"epoch": 1.4164413196322336,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004984404028148585,
|
|
"loss": 5.8533,
|
|
"mean_token_accuracy": 0.1614123612642288,
|
|
"num_tokens": 2941299.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 5.845489120483398,
|
|
"epoch": 1.4218496484586263,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004983895828427062,
|
|
"loss": 5.7794,
|
|
"mean_token_accuracy": 0.15733043551445008,
|
|
"num_tokens": 2952283.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.1262627124786375,
|
|
"epoch": 1.427257977285019,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004983379510722935,
|
|
"loss": 6.157,
|
|
"mean_token_accuracy": 0.15203749984502793,
|
|
"num_tokens": 2962241.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 5.937980365753174,
|
|
"epoch": 1.4326663061114115,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004982855076912506,
|
|
"loss": 5.9127,
|
|
"mean_token_accuracy": 0.15764199197292328,
|
|
"num_tokens": 2973408.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 5.948457860946656,
|
|
"epoch": 1.4380746349378042,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004982322528901568,
|
|
"loss": 5.9232,
|
|
"mean_token_accuracy": 0.15962736383080484,
|
|
"num_tokens": 2985877.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 5.9929845333099365,
|
|
"epoch": 1.4434829637641968,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00049817818686254,
|
|
"loss": 5.9264,
|
|
"mean_token_accuracy": 0.157138629257679,
|
|
"num_tokens": 2996935.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 5.920670747756958,
|
|
"epoch": 1.4488912925905895,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004981233098048765,
|
|
"loss": 5.803,
|
|
"mean_token_accuracy": 0.15964050889015197,
|
|
"num_tokens": 3007932.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 5.968259859085083,
|
|
"epoch": 1.454299621416982,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004980676219165894,
|
|
"loss": 5.9104,
|
|
"mean_token_accuracy": 0.15741660445928574,
|
|
"num_tokens": 3019368.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.012629222869873,
|
|
"epoch": 1.4597079502433747,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000498011123400049,
|
|
"loss": 5.9693,
|
|
"mean_token_accuracy": 0.15534972846508027,
|
|
"num_tokens": 3029787.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 5.871347856521607,
|
|
"epoch": 1.4651162790697674,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004979538144605707,
|
|
"loss": 5.9273,
|
|
"mean_token_accuracy": 0.1607024699449539,
|
|
"num_tokens": 3042100.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 5.959978723526001,
|
|
"epoch": 1.47052460789616,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004978956953064154,
|
|
"loss": 5.947,
|
|
"mean_token_accuracy": 0.15917170345783233,
|
|
"num_tokens": 3054160.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 5.946196985244751,
|
|
"epoch": 1.4759329367225527,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004978367661487883,
|
|
"loss": 5.8879,
|
|
"mean_token_accuracy": 0.16079033613204957,
|
|
"num_tokens": 3065401.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 5.949586677551269,
|
|
"epoch": 1.4813412655489453,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004977770272018379,
|
|
"loss": 5.8341,
|
|
"mean_token_accuracy": 0.15969335436820983,
|
|
"num_tokens": 3076137.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 5.953902244567871,
|
|
"epoch": 1.486749594375338,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000497716478682656,
|
|
"loss": 5.9652,
|
|
"mean_token_accuracy": 0.1619146853685379,
|
|
"num_tokens": 3087274.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 5.963164854049682,
|
|
"epoch": 1.4921579232017308,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004976551208112759,
|
|
"loss": 5.8387,
|
|
"mean_token_accuracy": 0.16379695534706115,
|
|
"num_tokens": 3098476.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 5.882930040359497,
|
|
"epoch": 1.4975662520281232,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004975929538106721,
|
|
"loss": 5.8617,
|
|
"mean_token_accuracy": 0.1588952735066414,
|
|
"num_tokens": 3110060.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 5.863749265670776,
|
|
"epoch": 1.502974580854516,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004975299779067598,
|
|
"loss": 5.7849,
|
|
"mean_token_accuracy": 0.15960211604833602,
|
|
"num_tokens": 3121032.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 5.929670190811157,
|
|
"epoch": 1.5083829096809085,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004974661933283936,
|
|
"loss": 5.8082,
|
|
"mean_token_accuracy": 0.1662907764315605,
|
|
"num_tokens": 3131999.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.761184024810791,
|
|
"epoch": 1.5137912385073014,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004974016003073667,
|
|
"loss": 5.7306,
|
|
"mean_token_accuracy": 0.16570134311914445,
|
|
"num_tokens": 3142693.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 5.827089548110962,
|
|
"epoch": 1.5191995673336938,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004973361990784104,
|
|
"loss": 5.7083,
|
|
"mean_token_accuracy": 0.15890194475650787,
|
|
"num_tokens": 3154179.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 5.835238361358643,
|
|
"epoch": 1.5246078961600866,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004972699898791927,
|
|
"loss": 5.784,
|
|
"mean_token_accuracy": 0.16352133005857467,
|
|
"num_tokens": 3165314.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.892254877090454,
|
|
"epoch": 1.530016224986479,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004972029729503182,
|
|
"loss": 5.9056,
|
|
"mean_token_accuracy": 0.1654028832912445,
|
|
"num_tokens": 3177646.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 5.9668666362762455,
|
|
"epoch": 1.535424553812872,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004971351485353265,
|
|
"loss": 5.8915,
|
|
"mean_token_accuracy": 0.1610996037721634,
|
|
"num_tokens": 3188401.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.7779381275177,
|
|
"epoch": 1.5408328826392643,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004970665168806917,
|
|
"loss": 5.7797,
|
|
"mean_token_accuracy": 0.15889904350042344,
|
|
"num_tokens": 3200206.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.902052307128907,
|
|
"epoch": 1.5462412114656572,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004969970782358214,
|
|
"loss": 5.8879,
|
|
"mean_token_accuracy": 0.16096998155117034,
|
|
"num_tokens": 3211981.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 5.868707990646362,
|
|
"epoch": 1.5516495402920496,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004969268328530557,
|
|
"loss": 5.9143,
|
|
"mean_token_accuracy": 0.15627580285072326,
|
|
"num_tokens": 3221997.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 5.980175733566284,
|
|
"epoch": 1.5570578691184425,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004968557809876666,
|
|
"loss": 5.8468,
|
|
"mean_token_accuracy": 0.1607530415058136,
|
|
"num_tokens": 3234333.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.825837516784668,
|
|
"epoch": 1.562466197944835,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004967839228978568,
|
|
"loss": 5.891,
|
|
"mean_token_accuracy": 0.153375443816185,
|
|
"num_tokens": 3245083.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 5.89825963973999,
|
|
"epoch": 1.5678745267712277,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004967112588447585,
|
|
"loss": 5.7312,
|
|
"mean_token_accuracy": 0.17310669720172883,
|
|
"num_tokens": 3255787.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.849927759170532,
|
|
"epoch": 1.5732828555976204,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004966377890924332,
|
|
"loss": 5.9076,
|
|
"mean_token_accuracy": 0.1595471516251564,
|
|
"num_tokens": 3267793.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.841161441802979,
|
|
"epoch": 1.578691184424013,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004965635139078704,
|
|
"loss": 5.7825,
|
|
"mean_token_accuracy": 0.16770703047513963,
|
|
"num_tokens": 3279100.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 5.936021137237549,
|
|
"epoch": 1.5840995132504057,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004964884335609857,
|
|
"loss": 5.9412,
|
|
"mean_token_accuracy": 0.16327747106552123,
|
|
"num_tokens": 3290251.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.7994856357574465,
|
|
"epoch": 1.5895078420767983,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004964125483246218,
|
|
"loss": 5.7974,
|
|
"mean_token_accuracy": 0.16179104298353195,
|
|
"num_tokens": 3301817.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.973921442031861,
|
|
"epoch": 1.594916170903191,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004963358584745457,
|
|
"loss": 5.8662,
|
|
"mean_token_accuracy": 0.1604505494236946,
|
|
"num_tokens": 3313435.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 5.851998615264892,
|
|
"epoch": 1.6003244997295836,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004962583642894482,
|
|
"loss": 5.9556,
|
|
"mean_token_accuracy": 0.15699442476034164,
|
|
"num_tokens": 3325489.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 5.955336093902588,
|
|
"epoch": 1.6057328285559762,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004961800660509437,
|
|
"loss": 5.8308,
|
|
"mean_token_accuracy": 0.16538330316543579,
|
|
"num_tokens": 3337030.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.784128093719483,
|
|
"epoch": 1.6111411573823688,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000496100964043568,
|
|
"loss": 5.8012,
|
|
"mean_token_accuracy": 0.16296399980783463,
|
|
"num_tokens": 3347766.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.802460527420044,
|
|
"epoch": 1.6165494862087615,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000496021058554778,
|
|
"loss": 5.7202,
|
|
"mean_token_accuracy": 0.1604635998606682,
|
|
"num_tokens": 3360379.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.804496765136719,
|
|
"epoch": 1.6219578150351541,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004959403498749505,
|
|
"loss": 5.8475,
|
|
"mean_token_accuracy": 0.16599454432725907,
|
|
"num_tokens": 3370962.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.6219578150351541,
|
|
"eval_entropy": 5.789530323852192,
|
|
"eval_loss": 6.099030017852783,
|
|
"eval_mean_token_accuracy": 0.16077508542009375,
|
|
"eval_num_tokens": 3370962.0,
|
|
"eval_runtime": 2.7354,
|
|
"eval_samples_per_second": 1284.982,
|
|
"eval_steps_per_second": 160.851,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.877967977523804,
|
|
"epoch": 1.6273661438615468,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004958588382973809,
|
|
"loss": 5.7604,
|
|
"mean_token_accuracy": 0.16643749475479125,
|
|
"num_tokens": 3382648.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.847539615631104,
|
|
"epoch": 1.6327744726879394,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004957765241182826,
|
|
"loss": 5.8303,
|
|
"mean_token_accuracy": 0.16527807861566543,
|
|
"num_tokens": 3393871.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.815026140213012,
|
|
"epoch": 1.638182801514332,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004956934076367854,
|
|
"loss": 5.7212,
|
|
"mean_token_accuracy": 0.1676459342241287,
|
|
"num_tokens": 3405148.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.797288084030152,
|
|
"epoch": 1.6435911303407247,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000495609489154935,
|
|
"loss": 5.7581,
|
|
"mean_token_accuracy": 0.1730334922671318,
|
|
"num_tokens": 3416647.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.852180814743042,
|
|
"epoch": 1.6489994591671173,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004955247689776911,
|
|
"loss": 5.8515,
|
|
"mean_token_accuracy": 0.161680169403553,
|
|
"num_tokens": 3428331.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 5.764764928817749,
|
|
"epoch": 1.65440778799351,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000495439247412927,
|
|
"loss": 5.7162,
|
|
"mean_token_accuracy": 0.1679590366780758,
|
|
"num_tokens": 3440683.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.8154102802276615,
|
|
"epoch": 1.6598161168199026,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004953529247714285,
|
|
"loss": 5.7171,
|
|
"mean_token_accuracy": 0.16858665347099305,
|
|
"num_tokens": 3452375.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.848917865753174,
|
|
"epoch": 1.6652244456462952,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004952658013668922,
|
|
"loss": 5.8674,
|
|
"mean_token_accuracy": 0.15662428140640258,
|
|
"num_tokens": 3463255.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.882652664184571,
|
|
"epoch": 1.670632774472688,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004951778775159246,
|
|
"loss": 5.8086,
|
|
"mean_token_accuracy": 0.15849490314722062,
|
|
"num_tokens": 3473642.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.77342963218689,
|
|
"epoch": 1.6760411032990805,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004950891535380414,
|
|
"loss": 5.7758,
|
|
"mean_token_accuracy": 0.16221534013748168,
|
|
"num_tokens": 3484877.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.80095362663269,
|
|
"epoch": 1.6814494321254734,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004949996297556656,
|
|
"loss": 5.6964,
|
|
"mean_token_accuracy": 0.17248961329460144,
|
|
"num_tokens": 3495500.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.7876993179321286,
|
|
"epoch": 1.6868577609518658,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004949093064941268,
|
|
"loss": 5.8128,
|
|
"mean_token_accuracy": 0.16499137580394746,
|
|
"num_tokens": 3508057.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.928809690475464,
|
|
"epoch": 1.6922660897782587,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004948181840816601,
|
|
"loss": 5.9887,
|
|
"mean_token_accuracy": 0.15604898929595948,
|
|
"num_tokens": 3521110.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.857229852676392,
|
|
"epoch": 1.697674418604651,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004947262628494045,
|
|
"loss": 5.9006,
|
|
"mean_token_accuracy": 0.16128652840852736,
|
|
"num_tokens": 3530692.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.850112962722778,
|
|
"epoch": 1.703082747431044,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004946335431314018,
|
|
"loss": 5.8404,
|
|
"mean_token_accuracy": 0.15459179729223252,
|
|
"num_tokens": 3542195.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.898218822479248,
|
|
"epoch": 1.7084910762574363,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004945400252645959,
|
|
"loss": 5.7096,
|
|
"mean_token_accuracy": 0.17165947258472442,
|
|
"num_tokens": 3552676.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.718589353561401,
|
|
"epoch": 1.7138994050838292,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004944457095888308,
|
|
"loss": 5.6904,
|
|
"mean_token_accuracy": 0.17027546912431718,
|
|
"num_tokens": 3563308.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.806781816482544,
|
|
"epoch": 1.7193077339102216,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00049435059644685,
|
|
"loss": 5.7115,
|
|
"mean_token_accuracy": 0.1697133257985115,
|
|
"num_tokens": 3574085.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.825162696838379,
|
|
"epoch": 1.7247160627366145,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004942546861842947,
|
|
"loss": 5.7919,
|
|
"mean_token_accuracy": 0.16102959215641022,
|
|
"num_tokens": 3585816.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.8270790576934814,
|
|
"epoch": 1.730124391563007,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004941579791497032,
|
|
"loss": 5.8447,
|
|
"mean_token_accuracy": 0.16419899761676787,
|
|
"num_tokens": 3597776.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.9654731273651125,
|
|
"epoch": 1.7355327203893998,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004940604756945091,
|
|
"loss": 5.8474,
|
|
"mean_token_accuracy": 0.16718958914279938,
|
|
"num_tokens": 3607991.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.776391935348511,
|
|
"epoch": 1.7409410492157922,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004939621761730399,
|
|
"loss": 5.7461,
|
|
"mean_token_accuracy": 0.17374723106622697,
|
|
"num_tokens": 3618204.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.817115354537964,
|
|
"epoch": 1.746349378042185,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004938630809425166,
|
|
"loss": 5.8782,
|
|
"mean_token_accuracy": 0.16300468146800995,
|
|
"num_tokens": 3629558.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.879113435745239,
|
|
"epoch": 1.7517577068685775,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004937631903630513,
|
|
"loss": 5.8046,
|
|
"mean_token_accuracy": 0.16308850198984146,
|
|
"num_tokens": 3640773.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.756259870529175,
|
|
"epoch": 1.7571660356949703,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004936625047976468,
|
|
"loss": 5.7143,
|
|
"mean_token_accuracy": 0.16870833039283753,
|
|
"num_tokens": 3652186.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.780659770965576,
|
|
"epoch": 1.7625743645213627,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004935610246121944,
|
|
"loss": 5.7601,
|
|
"mean_token_accuracy": 0.16812037974596022,
|
|
"num_tokens": 3663068.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.8441273212432865,
|
|
"epoch": 1.7679826933477556,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004934587501754736,
|
|
"loss": 5.9051,
|
|
"mean_token_accuracy": 0.16036613881587983,
|
|
"num_tokens": 3673803.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.9988672733306885,
|
|
"epoch": 1.7733910221741482,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004933556818591498,
|
|
"loss": 5.8948,
|
|
"mean_token_accuracy": 0.16162540018558502,
|
|
"num_tokens": 3685482.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.72188549041748,
|
|
"epoch": 1.7787993510005409,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004932518200377736,
|
|
"loss": 5.717,
|
|
"mean_token_accuracy": 0.1761845678091049,
|
|
"num_tokens": 3697641.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.782721376419067,
|
|
"epoch": 1.7842076798269335,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004931471650887789,
|
|
"loss": 5.7457,
|
|
"mean_token_accuracy": 0.1688130483031273,
|
|
"num_tokens": 3709560.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.927535963058472,
|
|
"epoch": 1.7896160086533262,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004930417173924822,
|
|
"loss": 5.8693,
|
|
"mean_token_accuracy": 0.15874227806925773,
|
|
"num_tokens": 3720422.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.676965141296387,
|
|
"epoch": 1.7950243374797188,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004929354773320806,
|
|
"loss": 5.6917,
|
|
"mean_token_accuracy": 0.16631999462842942,
|
|
"num_tokens": 3731104.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.819168472290039,
|
|
"epoch": 1.8004326663061114,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004928284452936506,
|
|
"loss": 5.8609,
|
|
"mean_token_accuracy": 0.15684734135866166,
|
|
"num_tokens": 3742766.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.824142551422119,
|
|
"epoch": 1.805840995132504,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004927206216661473,
|
|
"loss": 5.7334,
|
|
"mean_token_accuracy": 0.17028089016675949,
|
|
"num_tokens": 3753064.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.798613882064819,
|
|
"epoch": 1.8112493239588967,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004926120068414015,
|
|
"loss": 5.7855,
|
|
"mean_token_accuracy": 0.16478607952594757,
|
|
"num_tokens": 3765183.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.827127933502197,
|
|
"epoch": 1.8166576527852893,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004925026012141201,
|
|
"loss": 5.7692,
|
|
"mean_token_accuracy": 0.1576091930270195,
|
|
"num_tokens": 3777902.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.713541650772095,
|
|
"epoch": 1.822065981611682,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004923924051818832,
|
|
"loss": 5.7405,
|
|
"mean_token_accuracy": 0.1662287339568138,
|
|
"num_tokens": 3788461.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.7556291103363035,
|
|
"epoch": 1.8274743104380746,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004922814191451435,
|
|
"loss": 5.6702,
|
|
"mean_token_accuracy": 0.16689209192991256,
|
|
"num_tokens": 3799574.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.847048902511597,
|
|
"epoch": 1.8328826392644673,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004921696435072245,
|
|
"loss": 5.8296,
|
|
"mean_token_accuracy": 0.16467416435480117,
|
|
"num_tokens": 3811889.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.800088930130005,
|
|
"epoch": 1.83829096809086,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004920570786743191,
|
|
"loss": 5.7802,
|
|
"mean_token_accuracy": 0.16198689341545106,
|
|
"num_tokens": 3822949.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.858513307571411,
|
|
"epoch": 1.8436992969172525,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004919437250554882,
|
|
"loss": 5.8851,
|
|
"mean_token_accuracy": 0.15937444269657136,
|
|
"num_tokens": 3834734.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.825388336181641,
|
|
"epoch": 1.8491076257436452,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004918295830626592,
|
|
"loss": 5.7231,
|
|
"mean_token_accuracy": 0.172882080078125,
|
|
"num_tokens": 3845508.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.75558876991272,
|
|
"epoch": 1.8545159545700378,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004917146531106242,
|
|
"loss": 5.7204,
|
|
"mean_token_accuracy": 0.16671993881464003,
|
|
"num_tokens": 3856500.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.929794263839722,
|
|
"epoch": 1.8599242833964305,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004915989356170389,
|
|
"loss": 5.7999,
|
|
"mean_token_accuracy": 0.16467273533344268,
|
|
"num_tokens": 3867877.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.733937835693359,
|
|
"epoch": 1.865332612222823,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004914824310024213,
|
|
"loss": 5.7869,
|
|
"mean_token_accuracy": 0.16537229269742965,
|
|
"num_tokens": 3878613.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.780276107788086,
|
|
"epoch": 1.870740941049216,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004913651396901489,
|
|
"loss": 5.702,
|
|
"mean_token_accuracy": 0.17118038684129716,
|
|
"num_tokens": 3888418.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.778665542602539,
|
|
"epoch": 1.8761492698756084,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000491247062106459,
|
|
"loss": 5.7794,
|
|
"mean_token_accuracy": 0.16762736886739732,
|
|
"num_tokens": 3899359.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.900399255752563,
|
|
"epoch": 1.8815575987020012,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004911281986804455,
|
|
"loss": 5.8886,
|
|
"mean_token_accuracy": 0.1609322413802147,
|
|
"num_tokens": 3911568.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.783387899398804,
|
|
"epoch": 1.8869659275283936,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004910085498440587,
|
|
"loss": 5.824,
|
|
"mean_token_accuracy": 0.15804475843906401,
|
|
"num_tokens": 3921597.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.788153505325317,
|
|
"epoch": 1.8923742563547865,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004908881160321025,
|
|
"loss": 5.7118,
|
|
"mean_token_accuracy": 0.16248784065246583,
|
|
"num_tokens": 3933124.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.805125331878662,
|
|
"epoch": 1.897782585181179,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004907668976822338,
|
|
"loss": 5.7387,
|
|
"mean_token_accuracy": 0.16944270730018615,
|
|
"num_tokens": 3944388.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.750275135040283,
|
|
"epoch": 1.9031909140075718,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004906448952349603,
|
|
"loss": 5.7116,
|
|
"mean_token_accuracy": 0.16525028347969056,
|
|
"num_tokens": 3955024.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.631555461883545,
|
|
"epoch": 1.9085992428339642,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004905221091336392,
|
|
"loss": 5.7099,
|
|
"mean_token_accuracy": 0.17296146601438522,
|
|
"num_tokens": 3966029.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.750810003280639,
|
|
"epoch": 1.914007571660357,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004903985398244753,
|
|
"loss": 5.6643,
|
|
"mean_token_accuracy": 0.16801048219203948,
|
|
"num_tokens": 3976066.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.851806211471557,
|
|
"epoch": 1.9194159004867495,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004902741877565201,
|
|
"loss": 5.8495,
|
|
"mean_token_accuracy": 0.16151788383722304,
|
|
"num_tokens": 3986467.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.821902656555176,
|
|
"epoch": 1.9248242293131423,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000490149053381669,
|
|
"loss": 5.8311,
|
|
"mean_token_accuracy": 0.16564096361398697,
|
|
"num_tokens": 3998208.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.824001836776733,
|
|
"epoch": 1.9302325581395348,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004900231371546608,
|
|
"loss": 5.7755,
|
|
"mean_token_accuracy": 0.16438187509775162,
|
|
"num_tokens": 4008202.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.741191673278808,
|
|
"epoch": 1.9356408869659276,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004898964395330753,
|
|
"loss": 5.6195,
|
|
"mean_token_accuracy": 0.17555784583091735,
|
|
"num_tokens": 4019256.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.621674871444702,
|
|
"epoch": 1.94104921579232,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004897689609773318,
|
|
"loss": 5.6736,
|
|
"mean_token_accuracy": 0.16949178129434586,
|
|
"num_tokens": 4029738.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.854261255264282,
|
|
"epoch": 1.946457544618713,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004896407019506879,
|
|
"loss": 5.9036,
|
|
"mean_token_accuracy": 0.1552858531475067,
|
|
"num_tokens": 4042277.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.836483955383301,
|
|
"epoch": 1.9518658734451053,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004895116629192371,
|
|
"loss": 5.7497,
|
|
"mean_token_accuracy": 0.17201155722141265,
|
|
"num_tokens": 4054019.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.734428787231446,
|
|
"epoch": 1.9572742022714982,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004893818443519074,
|
|
"loss": 5.8097,
|
|
"mean_token_accuracy": 0.1687105417251587,
|
|
"num_tokens": 4065562.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.912402725219726,
|
|
"epoch": 1.9626825310978906,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048925124672046,
|
|
"loss": 5.8615,
|
|
"mean_token_accuracy": 0.16666115671396256,
|
|
"num_tokens": 4076617.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.791778564453125,
|
|
"epoch": 1.9680908599242835,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004891198704994868,
|
|
"loss": 5.6658,
|
|
"mean_token_accuracy": 0.17475759238004684,
|
|
"num_tokens": 4087191.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.729851293563843,
|
|
"epoch": 1.973499188750676,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004889877161664096,
|
|
"loss": 5.6999,
|
|
"mean_token_accuracy": 0.17070139050483704,
|
|
"num_tokens": 4099150.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.790151357650757,
|
|
"epoch": 1.9789075175770687,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004888547842014771,
|
|
"loss": 5.7146,
|
|
"mean_token_accuracy": 0.1693179652094841,
|
|
"num_tokens": 4110299.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.70378851890564,
|
|
"epoch": 1.9843158464034614,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004887210750877648,
|
|
"loss": 5.7418,
|
|
"mean_token_accuracy": 0.16693680733442307,
|
|
"num_tokens": 4121682.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.740220308303833,
|
|
"epoch": 1.989724175229854,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004885865893111716,
|
|
"loss": 5.6851,
|
|
"mean_token_accuracy": 0.16529171764850617,
|
|
"num_tokens": 4132914.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.805868530273438,
|
|
"epoch": 1.9951325040562466,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004884513273604195,
|
|
"loss": 5.779,
|
|
"mean_token_accuracy": 0.1664469823241234,
|
|
"num_tokens": 4144213.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.787704997592503,
|
|
"epoch": 2.0,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 0.0004883152897270502,
|
|
"loss": 5.804,
|
|
"mean_token_accuracy": 0.16674391428629556,
|
|
"num_tokens": 4153778.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.694442415237427,
|
|
"epoch": 2.005408328826393,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004881784769054253,
|
|
"loss": 5.3714,
|
|
"mean_token_accuracy": 0.17460792362689972,
|
|
"num_tokens": 4165536.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.842467737197876,
|
|
"epoch": 2.0108166576527853,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004880408893927226,
|
|
"loss": 5.4221,
|
|
"mean_token_accuracy": 0.16779832988977433,
|
|
"num_tokens": 4176182.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.570916748046875,
|
|
"epoch": 2.016224986479178,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004879025276889355,
|
|
"loss": 5.4388,
|
|
"mean_token_accuracy": 0.18602266013622284,
|
|
"num_tokens": 4187592.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.658016204833984,
|
|
"epoch": 2.0216333153055706,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048776339229687077,
|
|
"loss": 5.2193,
|
|
"mean_token_accuracy": 0.1936914801597595,
|
|
"num_tokens": 4198199.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.601340007781983,
|
|
"epoch": 2.0270416441319634,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048762348372214683,
|
|
"loss": 5.4708,
|
|
"mean_token_accuracy": 0.17636169642210006,
|
|
"num_tokens": 4208116.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.666546773910523,
|
|
"epoch": 2.032449972958356,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048748280247319165,
|
|
"loss": 5.2391,
|
|
"mean_token_accuracy": 0.18481917381286622,
|
|
"num_tokens": 4218569.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.635177087783814,
|
|
"epoch": 2.0378583017847487,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004873413490612412,
|
|
"loss": 5.4041,
|
|
"mean_token_accuracy": 0.18552399873733522,
|
|
"num_tokens": 4228785.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.657715606689453,
|
|
"epoch": 2.043266630611141,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048719912400033757,
|
|
"loss": 5.3771,
|
|
"mean_token_accuracy": 0.17886410504579545,
|
|
"num_tokens": 4238814.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.594145250320435,
|
|
"epoch": 2.048674959437534,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000487056127807327,
|
|
"loss": 5.3205,
|
|
"mean_token_accuracy": 0.17897071540355683,
|
|
"num_tokens": 4250544.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.656544637680054,
|
|
"epoch": 2.0540832882639264,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048691236100185795,
|
|
"loss": 5.4578,
|
|
"mean_token_accuracy": 0.17774118036031722,
|
|
"num_tokens": 4261806.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.734396934509277,
|
|
"epoch": 2.0594916170903192,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004867678241063792,
|
|
"loss": 5.4066,
|
|
"mean_token_accuracy": 0.1780041590332985,
|
|
"num_tokens": 4272472.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.631393194198608,
|
|
"epoch": 2.0648999459167117,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048662251764613833,
|
|
"loss": 5.3667,
|
|
"mean_token_accuracy": 0.17942027896642684,
|
|
"num_tokens": 4283758.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.631641101837158,
|
|
"epoch": 2.0703082747431045,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004864764421491793,
|
|
"loss": 5.4985,
|
|
"mean_token_accuracy": 0.1746978171169758,
|
|
"num_tokens": 4295930.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.763692808151245,
|
|
"epoch": 2.075716603569497,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004863295981463408,
|
|
"loss": 5.4092,
|
|
"mean_token_accuracy": 0.18891796916723252,
|
|
"num_tokens": 4306711.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.587791013717651,
|
|
"epoch": 2.08112493239589,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004861819861712543,
|
|
"loss": 5.3999,
|
|
"mean_token_accuracy": 0.1850850597023964,
|
|
"num_tokens": 4317694.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.719867706298828,
|
|
"epoch": 2.086533261222282,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048603360676034203,
|
|
"loss": 5.4896,
|
|
"mean_token_accuracy": 0.17167917191982268,
|
|
"num_tokens": 4329552.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.699314975738526,
|
|
"epoch": 2.091941590048675,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048588446045281505,
|
|
"loss": 5.4224,
|
|
"mean_token_accuracy": 0.18377942889928817,
|
|
"num_tokens": 4340479.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.493666553497315,
|
|
"epoch": 2.0973499188750675,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004857345477906714,
|
|
"loss": 5.3083,
|
|
"mean_token_accuracy": 0.18363658636808394,
|
|
"num_tokens": 4351147.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.502722930908203,
|
|
"epoch": 2.1027582477014604,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048558386931869417,
|
|
"loss": 5.3245,
|
|
"mean_token_accuracy": 0.18297460079193115,
|
|
"num_tokens": 4362962.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.623068714141846,
|
|
"epoch": 2.1081665765278528,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004854324255844492,
|
|
"loss": 5.4,
|
|
"mean_token_accuracy": 0.18296191245317459,
|
|
"num_tokens": 4373723.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.499787378311157,
|
|
"epoch": 2.1135749053542456,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048528021713828336,
|
|
"loss": 5.3819,
|
|
"mean_token_accuracy": 0.17632512450218202,
|
|
"num_tokens": 4383434.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.6350963592529295,
|
|
"epoch": 2.118983234180638,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004851272445333225,
|
|
"loss": 5.3411,
|
|
"mean_token_accuracy": 0.18811776787042617,
|
|
"num_tokens": 4394847.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.5932087898254395,
|
|
"epoch": 2.124391563007031,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048497350832546937,
|
|
"loss": 5.4447,
|
|
"mean_token_accuracy": 0.18534910231828688,
|
|
"num_tokens": 4406367.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.713639593124389,
|
|
"epoch": 2.1297998918334233,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004848190090734018,
|
|
"loss": 5.4309,
|
|
"mean_token_accuracy": 0.1798338308930397,
|
|
"num_tokens": 4417636.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.633783960342408,
|
|
"epoch": 2.135208220659816,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048466374733857035,
|
|
"loss": 5.4379,
|
|
"mean_token_accuracy": 0.18512897789478303,
|
|
"num_tokens": 4428473.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.554681491851807,
|
|
"epoch": 2.1406165494862086,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004845077236851966,
|
|
"loss": 5.3822,
|
|
"mean_token_accuracy": 0.1791573703289032,
|
|
"num_tokens": 4439780.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.619713020324707,
|
|
"epoch": 2.1460248783126015,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048435093868027076,
|
|
"loss": 5.3153,
|
|
"mean_token_accuracy": 0.1938255786895752,
|
|
"num_tokens": 4450929.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.559303426742554,
|
|
"epoch": 2.151433207138994,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048419339289354995,
|
|
"loss": 5.3649,
|
|
"mean_token_accuracy": 0.1821557641029358,
|
|
"num_tokens": 4462967.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.661104536056518,
|
|
"epoch": 2.1568415359653867,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000484035086897556,
|
|
"loss": 5.4587,
|
|
"mean_token_accuracy": 0.18290756195783614,
|
|
"num_tokens": 4473970.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.667714691162109,
|
|
"epoch": 2.162249864791779,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004838760212675732,
|
|
"loss": 5.4689,
|
|
"mean_token_accuracy": 0.17768498957157136,
|
|
"num_tokens": 4485660.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.162249864791779,
|
|
"eval_entropy": 5.625695644725453,
|
|
"eval_loss": 6.026589393615723,
|
|
"eval_mean_token_accuracy": 0.16454821371219375,
|
|
"eval_num_tokens": 4485660.0,
|
|
"eval_runtime": 2.5276,
|
|
"eval_samples_per_second": 1390.645,
|
|
"eval_steps_per_second": 174.078,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.60123724937439,
|
|
"epoch": 2.167658193618172,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004837161965816464,
|
|
"loss": 5.4393,
|
|
"mean_token_accuracy": 0.18373553901910783,
|
|
"num_tokens": 4497261.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.546935939788819,
|
|
"epoch": 2.1730665224445644,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048355561342057886,
|
|
"loss": 5.3232,
|
|
"mean_token_accuracy": 0.1893532618880272,
|
|
"num_tokens": 4508120.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.639460849761963,
|
|
"epoch": 2.1784748512709573,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048339427236793024,
|
|
"loss": 5.3877,
|
|
"mean_token_accuracy": 0.1875930905342102,
|
|
"num_tokens": 4519286.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.661156368255615,
|
|
"epoch": 2.1838831800973497,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004832321740100143,
|
|
"loss": 5.4865,
|
|
"mean_token_accuracy": 0.17418341487646102,
|
|
"num_tokens": 4531912.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.660031032562256,
|
|
"epoch": 2.1892915089237426,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048306931893589666,
|
|
"loss": 5.4839,
|
|
"mean_token_accuracy": 0.1804632067680359,
|
|
"num_tokens": 4544370.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.633778047561646,
|
|
"epoch": 2.194699837750135,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048290570773739327,
|
|
"loss": 5.4459,
|
|
"mean_token_accuracy": 0.17911728620529174,
|
|
"num_tokens": 4556013.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.568753671646118,
|
|
"epoch": 2.200108166576528,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048274134100906747,
|
|
"loss": 5.3823,
|
|
"mean_token_accuracy": 0.18302385360002518,
|
|
"num_tokens": 4566200.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.616315460205078,
|
|
"epoch": 2.2055164954029207,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048257621934822835,
|
|
"loss": 5.4262,
|
|
"mean_token_accuracy": 0.18204379975795745,
|
|
"num_tokens": 4576899.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.708587026596069,
|
|
"epoch": 2.210924824229313,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004824103433549285,
|
|
"loss": 5.44,
|
|
"mean_token_accuracy": 0.1758509173989296,
|
|
"num_tokens": 4588983.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.549923372268677,
|
|
"epoch": 2.2163331530557056,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004822437136319616,
|
|
"loss": 5.4014,
|
|
"mean_token_accuracy": 0.18566152900457383,
|
|
"num_tokens": 4599758.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.6263964653015135,
|
|
"epoch": 2.2217414818820984,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004820763307848604,
|
|
"loss": 5.4301,
|
|
"mean_token_accuracy": 0.17941709905862807,
|
|
"num_tokens": 4610927.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.580780792236328,
|
|
"epoch": 2.2271498107084913,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048190819542189456,
|
|
"loss": 5.475,
|
|
"mean_token_accuracy": 0.17878954857587814,
|
|
"num_tokens": 4623719.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.665354442596436,
|
|
"epoch": 2.2325581395348837,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048173930815406847,
|
|
"loss": 5.5174,
|
|
"mean_token_accuracy": 0.1811794623732567,
|
|
"num_tokens": 4635389.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.643967390060425,
|
|
"epoch": 2.2379664683612766,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004815696695951187,
|
|
"loss": 5.4645,
|
|
"mean_token_accuracy": 0.18017199039459228,
|
|
"num_tokens": 4645093.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.569951343536377,
|
|
"epoch": 2.243374797187669,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004813992803615123,
|
|
"loss": 5.4523,
|
|
"mean_token_accuracy": 0.17308327108621596,
|
|
"num_tokens": 4656490.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.69391450881958,
|
|
"epoch": 2.248783126014062,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00048122814107244414,
|
|
"loss": 5.4711,
|
|
"mean_token_accuracy": 0.18046900928020476,
|
|
"num_tokens": 4667603.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.5506998062133786,
|
|
"epoch": 2.2541914548404542,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048105625234983466,
|
|
"loss": 5.4145,
|
|
"mean_token_accuracy": 0.17359131425619126,
|
|
"num_tokens": 4678669.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.488923501968384,
|
|
"epoch": 2.259599783666847,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00048088361481832807,
|
|
"loss": 5.2131,
|
|
"mean_token_accuracy": 0.19622790962457656,
|
|
"num_tokens": 4689757.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.496966743469239,
|
|
"epoch": 2.2650081124932395,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048071022910528934,
|
|
"loss": 5.3875,
|
|
"mean_token_accuracy": 0.1851786270737648,
|
|
"num_tokens": 4699332.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.595490503311157,
|
|
"epoch": 2.2704164413196324,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048053609584080274,
|
|
"loss": 5.477,
|
|
"mean_token_accuracy": 0.18644167631864547,
|
|
"num_tokens": 4709485.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.613241672515869,
|
|
"epoch": 2.275824770146025,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048036121565766876,
|
|
"loss": 5.4225,
|
|
"mean_token_accuracy": 0.17837923467159272,
|
|
"num_tokens": 4720935.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.5089503765106205,
|
|
"epoch": 2.2812330989724177,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004801855891914026,
|
|
"loss": 5.3582,
|
|
"mean_token_accuracy": 0.17999012321233748,
|
|
"num_tokens": 4731219.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.645202684402466,
|
|
"epoch": 2.28664142779881,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048000921708023117,
|
|
"loss": 5.49,
|
|
"mean_token_accuracy": 0.17553784400224687,
|
|
"num_tokens": 4742637.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.554147529602051,
|
|
"epoch": 2.292049756625203,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004798320999650911,
|
|
"loss": 5.3839,
|
|
"mean_token_accuracy": 0.1789750948548317,
|
|
"num_tokens": 4753341.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.732188701629639,
|
|
"epoch": 2.2974580854515954,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047965423848962636,
|
|
"loss": 5.5243,
|
|
"mean_token_accuracy": 0.1844160109758377,
|
|
"num_tokens": 4765339.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.603096008300781,
|
|
"epoch": 2.302866414277988,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000479475633300186,
|
|
"loss": 5.3848,
|
|
"mean_token_accuracy": 0.18350827246904372,
|
|
"num_tokens": 4776149.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.595699453353882,
|
|
"epoch": 2.3082747431043806,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047929628504582173,
|
|
"loss": 5.4019,
|
|
"mean_token_accuracy": 0.18626316487789155,
|
|
"num_tokens": 4788120.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.664645481109619,
|
|
"epoch": 2.3136830719307735,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004791161943782856,
|
|
"loss": 5.4985,
|
|
"mean_token_accuracy": 0.17759274393320085,
|
|
"num_tokens": 4799922.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.481906747817993,
|
|
"epoch": 2.319091400757166,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047893536195202726,
|
|
"loss": 5.4369,
|
|
"mean_token_accuracy": 0.18070518374443054,
|
|
"num_tokens": 4810501.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.633908319473266,
|
|
"epoch": 2.3244997295835588,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004787537884241923,
|
|
"loss": 5.3853,
|
|
"mean_token_accuracy": 0.18269991874694824,
|
|
"num_tokens": 4820919.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.644043397903443,
|
|
"epoch": 2.329908058409951,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047857147445461943,
|
|
"loss": 5.462,
|
|
"mean_token_accuracy": 0.18295266181230546,
|
|
"num_tokens": 4833298.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.467016077041626,
|
|
"epoch": 2.335316387236344,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004783884207058379,
|
|
"loss": 5.3435,
|
|
"mean_token_accuracy": 0.17686130255460739,
|
|
"num_tokens": 4845199.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.607102012634277,
|
|
"epoch": 2.3407247160627365,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047820462784306555,
|
|
"loss": 5.3988,
|
|
"mean_token_accuracy": 0.18141285479068756,
|
|
"num_tokens": 4856549.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.580468463897705,
|
|
"epoch": 2.3461330448891293,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000478020096534206,
|
|
"loss": 5.4576,
|
|
"mean_token_accuracy": 0.17959757000207902,
|
|
"num_tokens": 4868446.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.621773099899292,
|
|
"epoch": 2.3515413737155217,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047783482744984644,
|
|
"loss": 5.5215,
|
|
"mean_token_accuracy": 0.17357867062091828,
|
|
"num_tokens": 4879047.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.654866790771484,
|
|
"epoch": 2.3569497025419146,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047764882126325527,
|
|
"loss": 5.3967,
|
|
"mean_token_accuracy": 0.176782064139843,
|
|
"num_tokens": 4891591.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.498101758956909,
|
|
"epoch": 2.362358031368307,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004774620786503793,
|
|
"loss": 5.4155,
|
|
"mean_token_accuracy": 0.1857382744550705,
|
|
"num_tokens": 4902710.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.601061916351318,
|
|
"epoch": 2.3677663601947,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004772746002898417,
|
|
"loss": 5.4762,
|
|
"mean_token_accuracy": 0.18371329605579376,
|
|
"num_tokens": 4914668.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.750937843322754,
|
|
"epoch": 2.3731746890210923,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047708638686293903,
|
|
"loss": 5.6104,
|
|
"mean_token_accuracy": 0.176163050532341,
|
|
"num_tokens": 4926684.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.644840288162231,
|
|
"epoch": 2.378583017847485,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047689743905363963,
|
|
"loss": 5.4811,
|
|
"mean_token_accuracy": 0.1778736650943756,
|
|
"num_tokens": 4938577.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.437067651748658,
|
|
"epoch": 2.383991346673878,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004767077575485801,
|
|
"loss": 5.3661,
|
|
"mean_token_accuracy": 0.19059835374355316,
|
|
"num_tokens": 4949070.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.601630640029907,
|
|
"epoch": 2.3893996755002704,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004765173430370635,
|
|
"loss": 5.4201,
|
|
"mean_token_accuracy": 0.18583499193191527,
|
|
"num_tokens": 4959558.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.528146743774414,
|
|
"epoch": 2.394808004326663,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004763261962110567,
|
|
"loss": 5.2853,
|
|
"mean_token_accuracy": 0.18994676917791367,
|
|
"num_tokens": 4970310.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.423049402236939,
|
|
"epoch": 2.4002163331530557,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004761343177651875,
|
|
"loss": 5.381,
|
|
"mean_token_accuracy": 0.17828835994005204,
|
|
"num_tokens": 4982300.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.596099805831909,
|
|
"epoch": 2.4056246619794486,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004759417083967428,
|
|
"loss": 5.4143,
|
|
"mean_token_accuracy": 0.18091550022363662,
|
|
"num_tokens": 4994425.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.56733250617981,
|
|
"epoch": 2.411032990805841,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047574836880566544,
|
|
"loss": 5.4326,
|
|
"mean_token_accuracy": 0.18720708936452865,
|
|
"num_tokens": 5005872.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.512912654876709,
|
|
"epoch": 2.4164413196322334,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004755542996945519,
|
|
"loss": 5.45,
|
|
"mean_token_accuracy": 0.1824631467461586,
|
|
"num_tokens": 5016903.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.627086544036866,
|
|
"epoch": 2.4218496484586263,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047535950176864976,
|
|
"loss": 5.4954,
|
|
"mean_token_accuracy": 0.1827078565955162,
|
|
"num_tokens": 5028407.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.653110122680664,
|
|
"epoch": 2.427257977285019,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047516397573585525,
|
|
"loss": 5.4939,
|
|
"mean_token_accuracy": 0.18210503607988357,
|
|
"num_tokens": 5039693.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.655086946487427,
|
|
"epoch": 2.4326663061114115,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047496772230671034,
|
|
"loss": 5.4396,
|
|
"mean_token_accuracy": 0.18439240157604217,
|
|
"num_tokens": 5051279.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.564061117172241,
|
|
"epoch": 2.4380746349378044,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004747707421944004,
|
|
"loss": 5.3716,
|
|
"mean_token_accuracy": 0.18237916231155396,
|
|
"num_tokens": 5062698.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.5757111549377445,
|
|
"epoch": 2.443482963764197,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004745730361147517,
|
|
"loss": 5.3818,
|
|
"mean_token_accuracy": 0.18398604840040206,
|
|
"num_tokens": 5074927.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.5645688533782955,
|
|
"epoch": 2.4488912925905897,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004743746047862286,
|
|
"loss": 5.5135,
|
|
"mean_token_accuracy": 0.16881413161754608,
|
|
"num_tokens": 5086298.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.651742744445801,
|
|
"epoch": 2.454299621416982,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004741754489299309,
|
|
"loss": 5.4424,
|
|
"mean_token_accuracy": 0.18015955686569213,
|
|
"num_tokens": 5098161.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.487890338897705,
|
|
"epoch": 2.459707950243375,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047397556926959155,
|
|
"loss": 5.4404,
|
|
"mean_token_accuracy": 0.18802594244480134,
|
|
"num_tokens": 5110743.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.689940977096557,
|
|
"epoch": 2.4651162790697674,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047377496653157374,
|
|
"loss": 5.4627,
|
|
"mean_token_accuracy": 0.18476583063602448,
|
|
"num_tokens": 5121838.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.540533208847046,
|
|
"epoch": 2.4705246078961602,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047357364144486806,
|
|
"loss": 5.4125,
|
|
"mean_token_accuracy": 0.17874770164489745,
|
|
"num_tokens": 5135047.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.513274002075195,
|
|
"epoch": 2.4759329367225527,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047337159474109057,
|
|
"loss": 5.397,
|
|
"mean_token_accuracy": 0.17887553572654724,
|
|
"num_tokens": 5146329.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.6418976306915285,
|
|
"epoch": 2.4813412655489455,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047316882715447934,
|
|
"loss": 5.442,
|
|
"mean_token_accuracy": 0.1812441051006317,
|
|
"num_tokens": 5157511.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.507776403427124,
|
|
"epoch": 2.486749594375338,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004729653394218922,
|
|
"loss": 5.4233,
|
|
"mean_token_accuracy": 0.1812361478805542,
|
|
"num_tokens": 5168456.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.696918058395386,
|
|
"epoch": 2.492157923201731,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047276113228280427,
|
|
"loss": 5.484,
|
|
"mean_token_accuracy": 0.17885934114456176,
|
|
"num_tokens": 5179294.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.59285306930542,
|
|
"epoch": 2.497566252028123,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004725562064793045,
|
|
"loss": 5.4385,
|
|
"mean_token_accuracy": 0.18642073422670363,
|
|
"num_tokens": 5190195.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.438157415390014,
|
|
"epoch": 2.502974580854516,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047235056275609397,
|
|
"loss": 5.3215,
|
|
"mean_token_accuracy": 0.18765312880277635,
|
|
"num_tokens": 5201289.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.543129110336304,
|
|
"epoch": 2.5083829096809085,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004721442018604822,
|
|
"loss": 5.3372,
|
|
"mean_token_accuracy": 0.18977320045232773,
|
|
"num_tokens": 5212295.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.458192539215088,
|
|
"epoch": 2.5137912385073014,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004719371245423853,
|
|
"loss": 5.3306,
|
|
"mean_token_accuracy": 0.19059140831232071,
|
|
"num_tokens": 5223144.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.487350845336914,
|
|
"epoch": 2.5191995673336938,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047172933155432273,
|
|
"loss": 5.3513,
|
|
"mean_token_accuracy": 0.1865290343761444,
|
|
"num_tokens": 5232840.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.573153400421143,
|
|
"epoch": 2.5246078961600866,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047152082365141466,
|
|
"loss": 5.4417,
|
|
"mean_token_accuracy": 0.18413045406341552,
|
|
"num_tokens": 5243796.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.621662044525147,
|
|
"epoch": 2.530016224986479,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004713116015913794,
|
|
"loss": 5.5396,
|
|
"mean_token_accuracy": 0.17397320568561553,
|
|
"num_tokens": 5255990.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.616555690765381,
|
|
"epoch": 2.535424553812872,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047110166613453016,
|
|
"loss": 5.49,
|
|
"mean_token_accuracy": 0.1783996805548668,
|
|
"num_tokens": 5268134.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.498351383209228,
|
|
"epoch": 2.5408328826392643,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047089101804377305,
|
|
"loss": 5.3353,
|
|
"mean_token_accuracy": 0.18645087331533433,
|
|
"num_tokens": 5278662.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.578673267364502,
|
|
"epoch": 2.546241211465657,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004706796580846037,
|
|
"loss": 5.4013,
|
|
"mean_token_accuracy": 0.18003640472888946,
|
|
"num_tokens": 5288382.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.556174230575562,
|
|
"epoch": 2.5516495402920496,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047046758702510465,
|
|
"loss": 5.4602,
|
|
"mean_token_accuracy": 0.17507169842720033,
|
|
"num_tokens": 5299769.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.649673128128052,
|
|
"epoch": 2.5570578691184425,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047025480563594264,
|
|
"loss": 5.3891,
|
|
"mean_token_accuracy": 0.18936282247304917,
|
|
"num_tokens": 5310662.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.485260820388794,
|
|
"epoch": 2.5624661979448353,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047004131469036576,
|
|
"loss": 5.3732,
|
|
"mean_token_accuracy": 0.17775198817253113,
|
|
"num_tokens": 5321785.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.567721748352051,
|
|
"epoch": 2.5678745267712277,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046982711496420057,
|
|
"loss": 5.4173,
|
|
"mean_token_accuracy": 0.17894906550645828,
|
|
"num_tokens": 5333557.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.568003463745117,
|
|
"epoch": 2.57328285559762,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046961220723584925,
|
|
"loss": 5.4826,
|
|
"mean_token_accuracy": 0.17998433262109756,
|
|
"num_tokens": 5344767.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.5262425422668455,
|
|
"epoch": 2.578691184424013,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046939659228628725,
|
|
"loss": 5.3969,
|
|
"mean_token_accuracy": 0.18416229337453843,
|
|
"num_tokens": 5356673.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.601033544540405,
|
|
"epoch": 2.584099513250406,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046918027089905956,
|
|
"loss": 5.5014,
|
|
"mean_token_accuracy": 0.17849208563566207,
|
|
"num_tokens": 5367296.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.492757177352905,
|
|
"epoch": 2.5895078420767983,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046896324386027873,
|
|
"loss": 5.3822,
|
|
"mean_token_accuracy": 0.18375635296106338,
|
|
"num_tokens": 5378081.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.656360864639282,
|
|
"epoch": 2.5949161709031907,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046874551195862145,
|
|
"loss": 5.4752,
|
|
"mean_token_accuracy": 0.178321735560894,
|
|
"num_tokens": 5390244.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.487719774246216,
|
|
"epoch": 2.6003244997295836,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000468527075985326,
|
|
"loss": 5.3328,
|
|
"mean_token_accuracy": 0.18906773924827575,
|
|
"num_tokens": 5400741.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.558295679092407,
|
|
"epoch": 2.6057328285559764,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046830793673418924,
|
|
"loss": 5.5083,
|
|
"mean_token_accuracy": 0.1851433753967285,
|
|
"num_tokens": 5411440.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.568044757843017,
|
|
"epoch": 2.611141157382369,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046808809500156355,
|
|
"loss": 5.3704,
|
|
"mean_token_accuracy": 0.1771426245570183,
|
|
"num_tokens": 5422472.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.4715736389160154,
|
|
"epoch": 2.6165494862087613,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004678675515863545,
|
|
"loss": 5.4606,
|
|
"mean_token_accuracy": 0.17648917734622954,
|
|
"num_tokens": 5433958.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.504480504989624,
|
|
"epoch": 2.621957815035154,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004676463072900173,
|
|
"loss": 5.3564,
|
|
"mean_token_accuracy": 0.1927761286497116,
|
|
"num_tokens": 5444343.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.501656198501587,
|
|
"epoch": 2.627366143861547,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00046742436291655415,
|
|
"loss": 5.3694,
|
|
"mean_token_accuracy": 0.1826394185423851,
|
|
"num_tokens": 5456260.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.591298246383667,
|
|
"epoch": 2.6327744726879394,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046720171927251155,
|
|
"loss": 5.4733,
|
|
"mean_token_accuracy": 0.18762189894914627,
|
|
"num_tokens": 5468281.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.393222665786743,
|
|
"epoch": 2.638182801514332,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004669783771669769,
|
|
"loss": 5.3231,
|
|
"mean_token_accuracy": 0.1976226106286049,
|
|
"num_tokens": 5480068.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.479925537109375,
|
|
"epoch": 2.6435911303407247,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000466754337411576,
|
|
"loss": 5.3197,
|
|
"mean_token_accuracy": 0.18385685235261917,
|
|
"num_tokens": 5490932.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.458782768249511,
|
|
"epoch": 2.6489994591671175,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046652960082046985,
|
|
"loss": 5.4136,
|
|
"mean_token_accuracy": 0.18229856789112092,
|
|
"num_tokens": 5501787.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.575427007675171,
|
|
"epoch": 2.65440778799351,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046630416821035164,
|
|
"loss": 5.3929,
|
|
"mean_token_accuracy": 0.1796463891863823,
|
|
"num_tokens": 5513297.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.510890626907349,
|
|
"epoch": 2.6598161168199024,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004660780404004441,
|
|
"loss": 5.3843,
|
|
"mean_token_accuracy": 0.1865953892469406,
|
|
"num_tokens": 5523797.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.436913394927979,
|
|
"epoch": 2.6652244456462952,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046585121821249623,
|
|
"loss": 5.4198,
|
|
"mean_token_accuracy": 0.18777260780334473,
|
|
"num_tokens": 5535808.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.638918256759643,
|
|
"epoch": 2.670632774472688,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046562370247078025,
|
|
"loss": 5.5181,
|
|
"mean_token_accuracy": 0.17214820235967637,
|
|
"num_tokens": 5548179.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.578511428833008,
|
|
"epoch": 2.6760411032990805,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000465395494002089,
|
|
"loss": 5.4903,
|
|
"mean_token_accuracy": 0.17955636084079743,
|
|
"num_tokens": 5559144.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.554292392730713,
|
|
"epoch": 2.6814494321254734,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004651665936357325,
|
|
"loss": 5.4074,
|
|
"mean_token_accuracy": 0.18264788389205933,
|
|
"num_tokens": 5571103.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.480365562438965,
|
|
"epoch": 2.686857760951866,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004649370022035354,
|
|
"loss": 5.3308,
|
|
"mean_token_accuracy": 0.1834966868162155,
|
|
"num_tokens": 5583239.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.415350341796875,
|
|
"epoch": 2.6922660897782587,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004647067205398332,
|
|
"loss": 5.3089,
|
|
"mean_token_accuracy": 0.18734012246131898,
|
|
"num_tokens": 5593602.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.621503257751465,
|
|
"epoch": 2.697674418604651,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046447574948147027,
|
|
"loss": 5.4706,
|
|
"mean_token_accuracy": 0.1750875771045685,
|
|
"num_tokens": 5604696.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.523394203186035,
|
|
"epoch": 2.703082747431044,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046424408986779587,
|
|
"loss": 5.4464,
|
|
"mean_token_accuracy": 0.17921075373888015,
|
|
"num_tokens": 5615668.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.703082747431044,
|
|
"eval_entropy": 5.505822956562042,
|
|
"eval_loss": 5.942599773406982,
|
|
"eval_mean_token_accuracy": 0.16984617078507488,
|
|
"eval_num_tokens": 5615668.0,
|
|
"eval_runtime": 2.5238,
|
|
"eval_samples_per_second": 1392.725,
|
|
"eval_steps_per_second": 174.338,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.535042238235474,
|
|
"epoch": 2.7084910762574363,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004640117425406615,
|
|
"loss": 5.3738,
|
|
"mean_token_accuracy": 0.18068106919527055,
|
|
"num_tokens": 5627344.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.487998056411743,
|
|
"epoch": 2.713899405083829,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000463778708344418,
|
|
"loss": 5.4028,
|
|
"mean_token_accuracy": 0.1800418332219124,
|
|
"num_tokens": 5637472.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.535551452636719,
|
|
"epoch": 2.7193077339102216,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046354498812591207,
|
|
"loss": 5.4057,
|
|
"mean_token_accuracy": 0.1849636197090149,
|
|
"num_tokens": 5647766.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.4707800388336185,
|
|
"epoch": 2.7247160627366145,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046331058273448364,
|
|
"loss": 5.3423,
|
|
"mean_token_accuracy": 0.18284157216548919,
|
|
"num_tokens": 5658343.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.497352409362793,
|
|
"epoch": 2.730124391563007,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004630754930219624,
|
|
"loss": 5.3272,
|
|
"mean_token_accuracy": 0.18483982384204864,
|
|
"num_tokens": 5669640.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.382404422760009,
|
|
"epoch": 2.7355327203893998,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004628397198426648,
|
|
"loss": 5.3152,
|
|
"mean_token_accuracy": 0.17685600370168686,
|
|
"num_tokens": 5682421.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.4589673519134525,
|
|
"epoch": 2.740941049215792,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046260326405339126,
|
|
"loss": 5.2853,
|
|
"mean_token_accuracy": 0.19155533462762833,
|
|
"num_tokens": 5692797.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.558311986923218,
|
|
"epoch": 2.746349378042185,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004623661265134226,
|
|
"loss": 5.4783,
|
|
"mean_token_accuracy": 0.18676115870475768,
|
|
"num_tokens": 5702779.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.5392842292785645,
|
|
"epoch": 2.7517577068685775,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004621283080845172,
|
|
"loss": 5.4478,
|
|
"mean_token_accuracy": 0.18404894173145295,
|
|
"num_tokens": 5714147.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.430927896499634,
|
|
"epoch": 2.7571660356949703,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00046188980963090783,
|
|
"loss": 5.2873,
|
|
"mean_token_accuracy": 0.19130903333425522,
|
|
"num_tokens": 5725718.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.423432731628418,
|
|
"epoch": 2.7625743645213627,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046165063201929825,
|
|
"loss": 5.3536,
|
|
"mean_token_accuracy": 0.19322678595781326,
|
|
"num_tokens": 5737441.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.588803148269653,
|
|
"epoch": 2.7679826933477556,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046141077611886074,
|
|
"loss": 5.4639,
|
|
"mean_token_accuracy": 0.1837845802307129,
|
|
"num_tokens": 5748735.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.593738031387329,
|
|
"epoch": 2.773391022174148,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004611702428012319,
|
|
"loss": 5.5198,
|
|
"mean_token_accuracy": 0.1718070924282074,
|
|
"num_tokens": 5760262.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.500187873840332,
|
|
"epoch": 2.778799351000541,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046092903294051057,
|
|
"loss": 5.4424,
|
|
"mean_token_accuracy": 0.18197216242551803,
|
|
"num_tokens": 5771909.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.433067703247071,
|
|
"epoch": 2.7842076798269337,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046068714741325385,
|
|
"loss": 5.3228,
|
|
"mean_token_accuracy": 0.18582267314195633,
|
|
"num_tokens": 5782028.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.618330669403076,
|
|
"epoch": 2.789616008653326,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004604445870984743,
|
|
"loss": 5.4593,
|
|
"mean_token_accuracy": 0.18109750002622604,
|
|
"num_tokens": 5793641.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.591123819351196,
|
|
"epoch": 2.7950243374797186,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004602013528776367,
|
|
"loss": 5.4157,
|
|
"mean_token_accuracy": 0.178664231300354,
|
|
"num_tokens": 5805392.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.533513879776001,
|
|
"epoch": 2.8004326663061114,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004599574456346548,
|
|
"loss": 5.3953,
|
|
"mean_token_accuracy": 0.18629478365182878,
|
|
"num_tokens": 5816362.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.46223521232605,
|
|
"epoch": 2.8058409951325043,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045971286625588796,
|
|
"loss": 5.3389,
|
|
"mean_token_accuracy": 0.18317433297634125,
|
|
"num_tokens": 5827518.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.594646692276001,
|
|
"epoch": 2.8112493239588967,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004594676156301383,
|
|
"loss": 5.4567,
|
|
"mean_token_accuracy": 0.18528510630130768,
|
|
"num_tokens": 5838455.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.504221820831299,
|
|
"epoch": 2.816657652785289,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045922169464864725,
|
|
"loss": 5.3472,
|
|
"mean_token_accuracy": 0.1895795449614525,
|
|
"num_tokens": 5850065.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.5160376071929935,
|
|
"epoch": 2.822065981611682,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000458975104205092,
|
|
"loss": 5.4604,
|
|
"mean_token_accuracy": 0.17686538994312287,
|
|
"num_tokens": 5860444.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.610097408294678,
|
|
"epoch": 2.827474310438075,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045872784519558304,
|
|
"loss": 5.4269,
|
|
"mean_token_accuracy": 0.1828258991241455,
|
|
"num_tokens": 5873984.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.531096410751343,
|
|
"epoch": 2.8328826392644673,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004584799185186599,
|
|
"loss": 5.4819,
|
|
"mean_token_accuracy": 0.18471675664186477,
|
|
"num_tokens": 5885807.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.440010833740234,
|
|
"epoch": 2.8382909680908597,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045823132507528903,
|
|
"loss": 5.346,
|
|
"mean_token_accuracy": 0.18770514726638793,
|
|
"num_tokens": 5897122.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.470519161224365,
|
|
"epoch": 2.8436992969172525,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004579820657688593,
|
|
"loss": 5.3507,
|
|
"mean_token_accuracy": 0.186607962846756,
|
|
"num_tokens": 5909189.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.558169364929199,
|
|
"epoch": 2.8491076257436454,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004577321415051798,
|
|
"loss": 5.4392,
|
|
"mean_token_accuracy": 0.17699794769287108,
|
|
"num_tokens": 5920135.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.545064067840576,
|
|
"epoch": 2.854515954570038,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045748155319247564,
|
|
"loss": 5.4361,
|
|
"mean_token_accuracy": 0.1826549470424652,
|
|
"num_tokens": 5932279.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.445204019546509,
|
|
"epoch": 2.8599242833964302,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004572303017413856,
|
|
"loss": 5.3579,
|
|
"mean_token_accuracy": 0.18405261188745498,
|
|
"num_tokens": 5942082.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.55396294593811,
|
|
"epoch": 2.865332612222823,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004569783880649578,
|
|
"loss": 5.3926,
|
|
"mean_token_accuracy": 0.18298255950212478,
|
|
"num_tokens": 5952856.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.401359415054321,
|
|
"epoch": 2.870740941049216,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004567258130786474,
|
|
"loss": 5.3087,
|
|
"mean_token_accuracy": 0.1933762788772583,
|
|
"num_tokens": 5963937.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.465329599380493,
|
|
"epoch": 2.8761492698756084,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045647257770031215,
|
|
"loss": 5.3907,
|
|
"mean_token_accuracy": 0.18456286638975145,
|
|
"num_tokens": 5974852.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.49418272972107,
|
|
"epoch": 2.8815575987020012,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045621868285021023,
|
|
"loss": 5.3851,
|
|
"mean_token_accuracy": 0.18949197679758073,
|
|
"num_tokens": 5985616.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.471716737747192,
|
|
"epoch": 2.8869659275283936,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00045596412945099615,
|
|
"loss": 5.3251,
|
|
"mean_token_accuracy": 0.1934232607483864,
|
|
"num_tokens": 5997199.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.4827837467193605,
|
|
"epoch": 2.8923742563547865,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045570891842771727,
|
|
"loss": 5.3969,
|
|
"mean_token_accuracy": 0.19077174365520477,
|
|
"num_tokens": 6008654.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.521725749969482,
|
|
"epoch": 2.897782585181179,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045545305070781135,
|
|
"loss": 5.4565,
|
|
"mean_token_accuracy": 0.18369679450988768,
|
|
"num_tokens": 6020662.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.515704917907715,
|
|
"epoch": 2.903190914007572,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045519652722110206,
|
|
"loss": 5.3305,
|
|
"mean_token_accuracy": 0.20163124799728394,
|
|
"num_tokens": 6031793.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.427778959274292,
|
|
"epoch": 2.908599242833964,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045493934889979655,
|
|
"loss": 5.3219,
|
|
"mean_token_accuracy": 0.18829119950532913,
|
|
"num_tokens": 6042181.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.359655141830444,
|
|
"epoch": 2.914007571660357,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045468151667848116,
|
|
"loss": 5.309,
|
|
"mean_token_accuracy": 0.18753138035535813,
|
|
"num_tokens": 6052774.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.519008588790894,
|
|
"epoch": 2.9194159004867495,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000454423031494119,
|
|
"loss": 5.2777,
|
|
"mean_token_accuracy": 0.1988011583685875,
|
|
"num_tokens": 6063273.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.4244945526123045,
|
|
"epoch": 2.9248242293131423,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004541638942860458,
|
|
"loss": 5.3297,
|
|
"mean_token_accuracy": 0.18525287955999375,
|
|
"num_tokens": 6075121.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.389618301391602,
|
|
"epoch": 2.9302325581395348,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004539041059959667,
|
|
"loss": 5.2741,
|
|
"mean_token_accuracy": 0.19131589233875274,
|
|
"num_tokens": 6085666.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.466776275634766,
|
|
"epoch": 2.9356408869659276,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045364366756795295,
|
|
"loss": 5.3106,
|
|
"mean_token_accuracy": 0.1948950007557869,
|
|
"num_tokens": 6096619.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.514159202575684,
|
|
"epoch": 2.94104921579232,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045338257994843845,
|
|
"loss": 5.4733,
|
|
"mean_token_accuracy": 0.1804106742143631,
|
|
"num_tokens": 6108800.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.5358567237854,
|
|
"epoch": 2.946457544618713,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004531208440862162,
|
|
"loss": 5.3803,
|
|
"mean_token_accuracy": 0.18469672948122023,
|
|
"num_tokens": 6119789.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.468766117095948,
|
|
"epoch": 2.9518658734451053,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045285846093243485,
|
|
"loss": 5.3746,
|
|
"mean_token_accuracy": 0.18340017944574355,
|
|
"num_tokens": 6131786.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.5707965850830075,
|
|
"epoch": 2.957274202271498,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045259543144059557,
|
|
"loss": 5.4558,
|
|
"mean_token_accuracy": 0.18204850554466248,
|
|
"num_tokens": 6143012.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.460539388656616,
|
|
"epoch": 2.9626825310978906,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004523317565665479,
|
|
"loss": 5.351,
|
|
"mean_token_accuracy": 0.18917624950408934,
|
|
"num_tokens": 6153972.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.475404167175293,
|
|
"epoch": 2.9680908599242835,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045206743726848716,
|
|
"loss": 5.4984,
|
|
"mean_token_accuracy": 0.1799850046634674,
|
|
"num_tokens": 6166202.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.419502925872803,
|
|
"epoch": 2.973499188750676,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004518024745069502,
|
|
"loss": 5.2372,
|
|
"mean_token_accuracy": 0.1963568478822708,
|
|
"num_tokens": 6177253.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.5437860012054445,
|
|
"epoch": 2.9789075175770687,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004515368692448123,
|
|
"loss": 5.5113,
|
|
"mean_token_accuracy": 0.1774690791964531,
|
|
"num_tokens": 6188857.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.574299097061157,
|
|
"epoch": 2.9843158464034616,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045127062244728353,
|
|
"loss": 5.4561,
|
|
"mean_token_accuracy": 0.17485650181770324,
|
|
"num_tokens": 6199945.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.463137292861939,
|
|
"epoch": 2.989724175229854,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004510037350819055,
|
|
"loss": 5.4438,
|
|
"mean_token_accuracy": 0.18347767144441604,
|
|
"num_tokens": 6209866.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.531537580490112,
|
|
"epoch": 2.9951325040562464,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045073620811854744,
|
|
"loss": 5.4131,
|
|
"mean_token_accuracy": 0.17728261500597,
|
|
"num_tokens": 6221636.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.48723840713501,
|
|
"epoch": 3.0,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 0.00045046804252940294,
|
|
"loss": 5.287,
|
|
"mean_token_accuracy": 0.18594140311082205,
|
|
"num_tokens": 6230667.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.438175535202026,
|
|
"epoch": 3.005408328826393,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045019923928898644,
|
|
"loss": 5.051,
|
|
"mean_token_accuracy": 0.2059848353266716,
|
|
"num_tokens": 6242242.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.5080938816070555,
|
|
"epoch": 3.0108166576527853,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004499297993741294,
|
|
"loss": 5.1042,
|
|
"mean_token_accuracy": 0.19538083225488662,
|
|
"num_tokens": 6253337.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.388686943054199,
|
|
"epoch": 3.016224986479178,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004496597237639773,
|
|
"loss": 4.971,
|
|
"mean_token_accuracy": 0.2036316141486168,
|
|
"num_tokens": 6264582.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.417944717407226,
|
|
"epoch": 3.0216333153055706,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044938901343998536,
|
|
"loss": 5.0126,
|
|
"mean_token_accuracy": 0.20922803729772568,
|
|
"num_tokens": 6276172.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.418192911148071,
|
|
"epoch": 3.0270416441319634,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004491176693859156,
|
|
"loss": 4.8813,
|
|
"mean_token_accuracy": 0.21112651377916336,
|
|
"num_tokens": 6286038.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.38726019859314,
|
|
"epoch": 3.032449972958356,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004488456925878329,
|
|
"loss": 4.9844,
|
|
"mean_token_accuracy": 0.20051523000001908,
|
|
"num_tokens": 6296546.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.367049503326416,
|
|
"epoch": 3.0378583017847487,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044857308403410163,
|
|
"loss": 5.068,
|
|
"mean_token_accuracy": 0.21011609137058257,
|
|
"num_tokens": 6308001.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.394755792617798,
|
|
"epoch": 3.043266630611141,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00044829984471538183,
|
|
"loss": 5.0801,
|
|
"mean_token_accuracy": 0.20128965228796006,
|
|
"num_tokens": 6320896.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.514834547042847,
|
|
"epoch": 3.048674959437534,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004480259756246258,
|
|
"loss": 5.12,
|
|
"mean_token_accuracy": 0.19838784784078597,
|
|
"num_tokens": 6332710.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.398886775970459,
|
|
"epoch": 3.0540832882639264,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044775147775707465,
|
|
"loss": 5.0013,
|
|
"mean_token_accuracy": 0.2039302572607994,
|
|
"num_tokens": 6343563.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.4193400859832765,
|
|
"epoch": 3.0594916170903192,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000447476352110254,
|
|
"loss": 5.1166,
|
|
"mean_token_accuracy": 0.20209966897964476,
|
|
"num_tokens": 6354064.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.384301424026489,
|
|
"epoch": 3.0648999459167117,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004472005996839712,
|
|
"loss": 5.0327,
|
|
"mean_token_accuracy": 0.20341972708702089,
|
|
"num_tokens": 6366551.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.351203441619873,
|
|
"epoch": 3.0703082747431045,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004469242214803112,
|
|
"loss": 4.9526,
|
|
"mean_token_accuracy": 0.20413549542427062,
|
|
"num_tokens": 6377086.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.3872153759002686,
|
|
"epoch": 3.075716603569497,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004466472185036331,
|
|
"loss": 5.0746,
|
|
"mean_token_accuracy": 0.20241950452327728,
|
|
"num_tokens": 6388795.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.402118730545044,
|
|
"epoch": 3.08112493239589,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004463695917605663,
|
|
"loss": 5.0522,
|
|
"mean_token_accuracy": 0.200471131503582,
|
|
"num_tokens": 6400174.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.365511322021485,
|
|
"epoch": 3.086533261222282,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000446091342260007,
|
|
"loss": 4.9981,
|
|
"mean_token_accuracy": 0.20258631110191344,
|
|
"num_tokens": 6411297.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.421813440322876,
|
|
"epoch": 3.091941590048675,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004458124710131145,
|
|
"loss": 5.0305,
|
|
"mean_token_accuracy": 0.2032555788755417,
|
|
"num_tokens": 6423487.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.502038955688477,
|
|
"epoch": 3.0973499188750675,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044553297903330766,
|
|
"loss": 5.2228,
|
|
"mean_token_accuracy": 0.19209739714860916,
|
|
"num_tokens": 6435460.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.4277064323425295,
|
|
"epoch": 3.1027582477014604,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044525286733626085,
|
|
"loss": 5.1559,
|
|
"mean_token_accuracy": 0.19844133853912355,
|
|
"num_tokens": 6445664.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.4236101627349855,
|
|
"epoch": 3.1081665765278528,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004449721369399008,
|
|
"loss": 5.033,
|
|
"mean_token_accuracy": 0.19993225932121278,
|
|
"num_tokens": 6457182.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.297450923919678,
|
|
"epoch": 3.1135749053542456,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044469078886440227,
|
|
"loss": 4.8812,
|
|
"mean_token_accuracy": 0.21529166549444198,
|
|
"num_tokens": 6468378.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.302643918991089,
|
|
"epoch": 3.118983234180638,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044440882413218484,
|
|
"loss": 5.1104,
|
|
"mean_token_accuracy": 0.1960832357406616,
|
|
"num_tokens": 6479625.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.368918561935425,
|
|
"epoch": 3.124391563007031,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044412624376790905,
|
|
"loss": 5.0236,
|
|
"mean_token_accuracy": 0.2056377574801445,
|
|
"num_tokens": 6490713.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.333490991592408,
|
|
"epoch": 3.1297998918334233,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004438430487984726,
|
|
"loss": 5.0854,
|
|
"mean_token_accuracy": 0.18910302668809892,
|
|
"num_tokens": 6502161.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.402835845947266,
|
|
"epoch": 3.135208220659816,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004435592402530066,
|
|
"loss": 4.981,
|
|
"mean_token_accuracy": 0.20249224305152894,
|
|
"num_tokens": 6513639.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.351930761337281,
|
|
"epoch": 3.1406165494862086,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000443274819162872,
|
|
"loss": 5.0097,
|
|
"mean_token_accuracy": 0.20482198148965836,
|
|
"num_tokens": 6524893.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.283976125717163,
|
|
"epoch": 3.1460248783126015,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004429897865616557,
|
|
"loss": 5.0058,
|
|
"mean_token_accuracy": 0.20499561876058578,
|
|
"num_tokens": 6535797.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.398959732055664,
|
|
"epoch": 3.151433207138994,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004427041434851668,
|
|
"loss": 5.0957,
|
|
"mean_token_accuracy": 0.19679532051086426,
|
|
"num_tokens": 6547117.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.44046196937561,
|
|
"epoch": 3.1568415359653867,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044241789097143293,
|
|
"loss": 5.1298,
|
|
"mean_token_accuracy": 0.20333695113658906,
|
|
"num_tokens": 6558771.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.323524475097656,
|
|
"epoch": 3.162249864791779,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044213103006069635,
|
|
"loss": 5.0933,
|
|
"mean_token_accuracy": 0.19716534912586212,
|
|
"num_tokens": 6569922.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.437791347503662,
|
|
"epoch": 3.167658193618172,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00044184356179541035,
|
|
"loss": 5.093,
|
|
"mean_token_accuracy": 0.20129518806934357,
|
|
"num_tokens": 6580561.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.3440930366516115,
|
|
"epoch": 3.1730665224445644,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004415554872202352,
|
|
"loss": 4.9166,
|
|
"mean_token_accuracy": 0.20625550150871277,
|
|
"num_tokens": 6591636.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.3810844898223875,
|
|
"epoch": 3.1784748512709573,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044126680738203466,
|
|
"loss": 5.1704,
|
|
"mean_token_accuracy": 0.19329363256692886,
|
|
"num_tokens": 6603736.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.475593709945679,
|
|
"epoch": 3.1838831800973497,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004409775233298718,
|
|
"loss": 5.087,
|
|
"mean_token_accuracy": 0.20025455951690674,
|
|
"num_tokens": 6615415.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.372749614715576,
|
|
"epoch": 3.1892915089237426,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004406876361150056,
|
|
"loss": 5.1276,
|
|
"mean_token_accuracy": 0.19769441187381745,
|
|
"num_tokens": 6625970.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.351458215713501,
|
|
"epoch": 3.194699837750135,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004403971467908869,
|
|
"loss": 5.0597,
|
|
"mean_token_accuracy": 0.19842777252197266,
|
|
"num_tokens": 6636326.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.421901702880859,
|
|
"epoch": 3.200108166576528,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004401060564131545,
|
|
"loss": 5.1314,
|
|
"mean_token_accuracy": 0.19361901730298997,
|
|
"num_tokens": 6646825.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.405378961563111,
|
|
"epoch": 3.2055164954029207,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004398143660396315,
|
|
"loss": 5.0955,
|
|
"mean_token_accuracy": 0.1982797846198082,
|
|
"num_tokens": 6657748.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.3214919090271,
|
|
"epoch": 3.210924824229313,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004395220767303215,
|
|
"loss": 4.9795,
|
|
"mean_token_accuracy": 0.21160258650779723,
|
|
"num_tokens": 6668422.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.376381921768188,
|
|
"epoch": 3.2163331530557056,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004392291895474045,
|
|
"loss": 5.0861,
|
|
"mean_token_accuracy": 0.19632923156023024,
|
|
"num_tokens": 6678898.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.436176586151123,
|
|
"epoch": 3.2217414818820984,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004389357055552331,
|
|
"loss": 5.124,
|
|
"mean_token_accuracy": 0.20098739862442017,
|
|
"num_tokens": 6688691.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.310462713241577,
|
|
"epoch": 3.2271498107084913,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00043864162582032897,
|
|
"loss": 5.01,
|
|
"mean_token_accuracy": 0.21020882278680803,
|
|
"num_tokens": 6700279.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.36355152130127,
|
|
"epoch": 3.2325581395348837,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004383469514113784,
|
|
"loss": 4.9687,
|
|
"mean_token_accuracy": 0.20986747443675996,
|
|
"num_tokens": 6710440.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.32545657157898,
|
|
"epoch": 3.2379664683612766,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004380516833992289,
|
|
"loss": 5.1212,
|
|
"mean_token_accuracy": 0.19812442511320114,
|
|
"num_tokens": 6722347.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.413011598587036,
|
|
"epoch": 3.243374797187669,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000437755822856885,
|
|
"loss": 5.1494,
|
|
"mean_token_accuracy": 0.19680777937173843,
|
|
"num_tokens": 6732936.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.243374797187669,
|
|
"eval_entropy": 5.348440836776387,
|
|
"eval_loss": 5.931386470794678,
|
|
"eval_mean_token_accuracy": 0.17230836965821006,
|
|
"eval_num_tokens": 6732936.0,
|
|
"eval_runtime": 2.7463,
|
|
"eval_samples_per_second": 1279.882,
|
|
"eval_steps_per_second": 160.213,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9240,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1.0532106150912e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|