117c2fdc6a
Model: fpadovani/eng-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
2057 lines
54 KiB
JSON
2057 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.0811249323958896,
|
|
"eval_steps": 500,
|
|
"global_step": 1000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 3492617103360000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|