20064 lines
549 KiB
JSON
20064 lines
549 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 9.990418557740798,
|
|
"eval_steps": 500,
|
|
"global_step": 9910,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691978168487548,
|
|
"epoch": 0.005042864346949067,
|
|
"grad_norm": 12.8125,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8351,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 11445.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691911983489991,
|
|
"epoch": 0.010085728693898134,
|
|
"grad_norm": 13.1875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7658,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 22095.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.691094779968262,
|
|
"epoch": 0.015128593040847202,
|
|
"grad_norm": 8.875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5389,
|
|
"mean_token_accuracy": 0.01894580915104598,
|
|
"num_tokens": 33847.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.681381797790527,
|
|
"epoch": 0.020171457387796268,
|
|
"grad_norm": 5.8125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.2545,
|
|
"mean_token_accuracy": 0.036951894313097,
|
|
"num_tokens": 44495.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.639113235473634,
|
|
"epoch": 0.025214321734745335,
|
|
"grad_norm": 4.0,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.9884,
|
|
"mean_token_accuracy": 0.038632451370358464,
|
|
"num_tokens": 55298.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.603860855102539,
|
|
"epoch": 0.030257186081694403,
|
|
"grad_norm": 3.21875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.8047,
|
|
"mean_token_accuracy": 0.04054084848612547,
|
|
"num_tokens": 66391.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.591868877410889,
|
|
"epoch": 0.03530005042864347,
|
|
"grad_norm": 2.65625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.698,
|
|
"mean_token_accuracy": 0.04207097329199314,
|
|
"num_tokens": 77624.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.584535121917725,
|
|
"epoch": 0.040342914775592535,
|
|
"grad_norm": 2.546875,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.678,
|
|
"mean_token_accuracy": 0.044261034950613974,
|
|
"num_tokens": 89426.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.562712478637696,
|
|
"epoch": 0.0453857791225416,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.6315,
|
|
"mean_token_accuracy": 0.04177976287901401,
|
|
"num_tokens": 100252.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.594598770141602,
|
|
"epoch": 0.05042864346949067,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.6087,
|
|
"mean_token_accuracy": 0.03483491390943527,
|
|
"num_tokens": 110578.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.596640396118165,
|
|
"epoch": 0.05547150781643974,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.5875,
|
|
"mean_token_accuracy": 0.03704058732837438,
|
|
"num_tokens": 121631.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.56342658996582,
|
|
"epoch": 0.060514372163388806,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.466,
|
|
"mean_token_accuracy": 0.05011895261704922,
|
|
"num_tokens": 134254.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.496850776672364,
|
|
"epoch": 0.06555723651033787,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.4511,
|
|
"mean_token_accuracy": 0.04732503369450569,
|
|
"num_tokens": 144928.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.411289405822753,
|
|
"epoch": 0.07060010085728693,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.2913,
|
|
"mean_token_accuracy": 0.04968489818274975,
|
|
"num_tokens": 155518.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.424620437622071,
|
|
"epoch": 0.07564296520423601,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.1933,
|
|
"mean_token_accuracy": 0.05173242129385471,
|
|
"num_tokens": 165850.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.444954109191894,
|
|
"epoch": 0.08068582955118507,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.1766,
|
|
"mean_token_accuracy": 0.04967899359762669,
|
|
"num_tokens": 176787.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.427998161315918,
|
|
"epoch": 0.08572869389813415,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.0647,
|
|
"mean_token_accuracy": 0.04877705723047256,
|
|
"num_tokens": 188345.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.456082916259765,
|
|
"epoch": 0.0907715582450832,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.0037,
|
|
"mean_token_accuracy": 0.04692690074443817,
|
|
"num_tokens": 199315.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.339052677154541,
|
|
"epoch": 0.09581442259203228,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.8109,
|
|
"mean_token_accuracy": 0.05837383419275284,
|
|
"num_tokens": 209627.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.241901874542236,
|
|
"epoch": 0.10085728693898134,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.7101,
|
|
"mean_token_accuracy": 0.0651729203760624,
|
|
"num_tokens": 220872.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.15722131729126,
|
|
"epoch": 0.1059001512859304,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.5657,
|
|
"mean_token_accuracy": 0.06282355673611165,
|
|
"num_tokens": 233272.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.10981206893921,
|
|
"epoch": 0.11094301563287948,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.4933,
|
|
"mean_token_accuracy": 0.06420289240777492,
|
|
"num_tokens": 243529.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.059538745880127,
|
|
"epoch": 0.11598587997982854,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.3948,
|
|
"mean_token_accuracy": 0.05781678333878517,
|
|
"num_tokens": 254942.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.973904132843018,
|
|
"epoch": 0.12102874432677761,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.2972,
|
|
"mean_token_accuracy": 0.06580851301550865,
|
|
"num_tokens": 266552.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.73906192779541,
|
|
"epoch": 0.1260716086737267,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.1497,
|
|
"mean_token_accuracy": 0.06753928288817405,
|
|
"num_tokens": 277906.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.731296634674072,
|
|
"epoch": 0.13111447302067575,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.0309,
|
|
"mean_token_accuracy": 0.06970244199037552,
|
|
"num_tokens": 289079.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.517922592163085,
|
|
"epoch": 0.1361573373676248,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.9818,
|
|
"mean_token_accuracy": 0.06449979990720749,
|
|
"num_tokens": 300107.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.309736728668213,
|
|
"epoch": 0.14120020171457387,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.8019,
|
|
"mean_token_accuracy": 0.06583718433976174,
|
|
"num_tokens": 310779.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.122577667236328,
|
|
"epoch": 0.14624306606152296,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.7857,
|
|
"mean_token_accuracy": 0.0686340156942606,
|
|
"num_tokens": 323478.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.875781536102295,
|
|
"epoch": 0.15128593040847202,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.662,
|
|
"mean_token_accuracy": 0.07218315117061139,
|
|
"num_tokens": 335255.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.774222087860107,
|
|
"epoch": 0.15632879475542108,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.5823,
|
|
"mean_token_accuracy": 0.07090610340237617,
|
|
"num_tokens": 344824.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.562890911102295,
|
|
"epoch": 0.16137165910237014,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.5779,
|
|
"mean_token_accuracy": 0.06917917579412461,
|
|
"num_tokens": 356283.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.52785816192627,
|
|
"epoch": 0.1664145234493192,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.5443,
|
|
"mean_token_accuracy": 0.07162331864237785,
|
|
"num_tokens": 368098.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.293674755096436,
|
|
"epoch": 0.1714573877962683,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.4885,
|
|
"mean_token_accuracy": 0.07210201621055604,
|
|
"num_tokens": 380948.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.206411838531494,
|
|
"epoch": 0.17650025214321735,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.4311,
|
|
"mean_token_accuracy": 0.07011085003614426,
|
|
"num_tokens": 390392.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.1397864818573,
|
|
"epoch": 0.1815431164901664,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.4686,
|
|
"mean_token_accuracy": 0.07223408669233322,
|
|
"num_tokens": 400328.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.180228519439698,
|
|
"epoch": 0.18658598083711547,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.4529,
|
|
"mean_token_accuracy": 0.07169274277985097,
|
|
"num_tokens": 410909.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.057634353637695,
|
|
"epoch": 0.19162884518406456,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.4423,
|
|
"mean_token_accuracy": 0.07097707688808441,
|
|
"num_tokens": 422726.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.975717973709107,
|
|
"epoch": 0.19667170953101362,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.3252,
|
|
"mean_token_accuracy": 0.07745771333575249,
|
|
"num_tokens": 433376.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.932044219970703,
|
|
"epoch": 0.20171457387796268,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.4045,
|
|
"mean_token_accuracy": 0.06969214789569378,
|
|
"num_tokens": 443940.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.979291915893555,
|
|
"epoch": 0.20675743822491174,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.2735,
|
|
"mean_token_accuracy": 0.07643608674407006,
|
|
"num_tokens": 455016.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.8824797630310055,
|
|
"epoch": 0.2118003025718608,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.3292,
|
|
"mean_token_accuracy": 0.07622824087738991,
|
|
"num_tokens": 466380.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.831804847717285,
|
|
"epoch": 0.2168431669188099,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.1783,
|
|
"mean_token_accuracy": 0.08013702630996704,
|
|
"num_tokens": 476915.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.80485315322876,
|
|
"epoch": 0.22188603126575895,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.2725,
|
|
"mean_token_accuracy": 0.0824611522257328,
|
|
"num_tokens": 488924.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.809728908538818,
|
|
"epoch": 0.22692889561270801,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.2162,
|
|
"mean_token_accuracy": 0.07684922367334365,
|
|
"num_tokens": 498965.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.819616842269897,
|
|
"epoch": 0.23197175995965708,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.235,
|
|
"mean_token_accuracy": 0.0765653271228075,
|
|
"num_tokens": 509755.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.771869659423828,
|
|
"epoch": 0.23701462430660616,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.2042,
|
|
"mean_token_accuracy": 0.0763448379933834,
|
|
"num_tokens": 521221.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.7396564960479735,
|
|
"epoch": 0.24205748865355523,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.3172,
|
|
"mean_token_accuracy": 0.07263238392770291,
|
|
"num_tokens": 533595.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.770624494552612,
|
|
"epoch": 0.24710035300050429,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.2608,
|
|
"mean_token_accuracy": 0.07456091418862343,
|
|
"num_tokens": 543672.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.743118524551392,
|
|
"epoch": 0.2521432173474534,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.127,
|
|
"mean_token_accuracy": 0.0765512753278017,
|
|
"num_tokens": 554906.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.675862503051758,
|
|
"epoch": 0.25718608169440244,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.2259,
|
|
"mean_token_accuracy": 0.0814048271626234,
|
|
"num_tokens": 566419.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.71265172958374,
|
|
"epoch": 0.2622289460413515,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.1691,
|
|
"mean_token_accuracy": 0.07766851969063282,
|
|
"num_tokens": 578848.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.5628129005432125,
|
|
"epoch": 0.26727181038830056,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.1197,
|
|
"mean_token_accuracy": 0.07973449081182479,
|
|
"num_tokens": 590601.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.681688404083252,
|
|
"epoch": 0.2723146747352496,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.1774,
|
|
"mean_token_accuracy": 0.07436439543962478,
|
|
"num_tokens": 602021.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.616954660415649,
|
|
"epoch": 0.2773575390821987,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.2583,
|
|
"mean_token_accuracy": 0.07453522607684135,
|
|
"num_tokens": 613966.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.710917282104492,
|
|
"epoch": 0.28240040342914774,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.0984,
|
|
"mean_token_accuracy": 0.07756517678499222,
|
|
"num_tokens": 624851.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.546781873703003,
|
|
"epoch": 0.2874432677760968,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.1915,
|
|
"mean_token_accuracy": 0.07566507011651993,
|
|
"num_tokens": 635580.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.662804460525512,
|
|
"epoch": 0.2924861321230459,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.2473,
|
|
"mean_token_accuracy": 0.07517430521547794,
|
|
"num_tokens": 646927.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.587956047058105,
|
|
"epoch": 0.297528996469995,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.0551,
|
|
"mean_token_accuracy": 0.08050617948174477,
|
|
"num_tokens": 658116.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.642880249023437,
|
|
"epoch": 0.30257186081694404,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.0584,
|
|
"mean_token_accuracy": 0.07990356534719467,
|
|
"num_tokens": 669054.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.5717206478118895,
|
|
"epoch": 0.3076147251638931,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.1075,
|
|
"mean_token_accuracy": 0.0826454609632492,
|
|
"num_tokens": 680439.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.453785753250122,
|
|
"epoch": 0.31265758951084216,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.0661,
|
|
"mean_token_accuracy": 0.08129210546612739,
|
|
"num_tokens": 690909.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.5011971950531,
|
|
"epoch": 0.3177004538577912,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.048,
|
|
"mean_token_accuracy": 0.07969367988407612,
|
|
"num_tokens": 701744.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.467720365524292,
|
|
"epoch": 0.3227433182047403,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.0817,
|
|
"mean_token_accuracy": 0.08535065315663815,
|
|
"num_tokens": 712268.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.59487681388855,
|
|
"epoch": 0.32778618255168934,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.1079,
|
|
"mean_token_accuracy": 0.07844235375523567,
|
|
"num_tokens": 722978.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.482430076599121,
|
|
"epoch": 0.3328290468986384,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.1011,
|
|
"mean_token_accuracy": 0.08162309601902962,
|
|
"num_tokens": 734278.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.5462571144104,
|
|
"epoch": 0.3378719112455875,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.0141,
|
|
"mean_token_accuracy": 0.0776023030281067,
|
|
"num_tokens": 746109.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.374982213973999,
|
|
"epoch": 0.3429147755925366,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.9356,
|
|
"mean_token_accuracy": 0.08509711548686028,
|
|
"num_tokens": 757519.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.443374300003052,
|
|
"epoch": 0.34795763993948564,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.9149,
|
|
"mean_token_accuracy": 0.08727683797478676,
|
|
"num_tokens": 767665.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.435840845108032,
|
|
"epoch": 0.3530005042864347,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.9537,
|
|
"mean_token_accuracy": 0.08823386654257774,
|
|
"num_tokens": 778880.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.310178327560425,
|
|
"epoch": 0.35804336863338376,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.9559,
|
|
"mean_token_accuracy": 0.09013235792517663,
|
|
"num_tokens": 790360.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.387903451919556,
|
|
"epoch": 0.3630862329803328,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.9182,
|
|
"mean_token_accuracy": 0.08709552362561226,
|
|
"num_tokens": 800889.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.419098663330078,
|
|
"epoch": 0.3681290973272819,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.9171,
|
|
"mean_token_accuracy": 0.08571921288967133,
|
|
"num_tokens": 811414.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.240619897842407,
|
|
"epoch": 0.37317196167423095,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.9305,
|
|
"mean_token_accuracy": 0.0850242905318737,
|
|
"num_tokens": 822822.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.271744012832642,
|
|
"epoch": 0.37821482602118,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.829,
|
|
"mean_token_accuracy": 0.09168825820088386,
|
|
"num_tokens": 833145.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.2786908626556395,
|
|
"epoch": 0.3832576903681291,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.9405,
|
|
"mean_token_accuracy": 0.0836037501692772,
|
|
"num_tokens": 844015.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.349726581573487,
|
|
"epoch": 0.3883005547150782,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.8706,
|
|
"mean_token_accuracy": 0.08842808604240418,
|
|
"num_tokens": 855806.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.209612941741943,
|
|
"epoch": 0.39334341906202724,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.7935,
|
|
"mean_token_accuracy": 0.08768922984600067,
|
|
"num_tokens": 865750.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.334837818145752,
|
|
"epoch": 0.3983862834089763,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.9248,
|
|
"mean_token_accuracy": 0.08573650568723679,
|
|
"num_tokens": 877630.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.157733392715454,
|
|
"epoch": 0.40342914775592537,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.8054,
|
|
"mean_token_accuracy": 0.09393974989652634,
|
|
"num_tokens": 888667.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.3036473274230955,
|
|
"epoch": 0.4084720121028744,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.8838,
|
|
"mean_token_accuracy": 0.08392557948827743,
|
|
"num_tokens": 900989.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.183315801620483,
|
|
"epoch": 0.4135148764498235,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.7766,
|
|
"mean_token_accuracy": 0.08522050082683563,
|
|
"num_tokens": 911479.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.263188695907592,
|
|
"epoch": 0.41855774079677255,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.7898,
|
|
"mean_token_accuracy": 0.08779488950967788,
|
|
"num_tokens": 922434.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.146833324432373,
|
|
"epoch": 0.4236006051437216,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.7957,
|
|
"mean_token_accuracy": 0.08833592608571053,
|
|
"num_tokens": 933405.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.173463964462281,
|
|
"epoch": 0.4286434694906707,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.7996,
|
|
"mean_token_accuracy": 0.0885507345199585,
|
|
"num_tokens": 943567.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.099128866195679,
|
|
"epoch": 0.4336863338376198,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.8177,
|
|
"mean_token_accuracy": 0.08491698876023293,
|
|
"num_tokens": 955123.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.277088403701782,
|
|
"epoch": 0.43872919818456885,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.8615,
|
|
"mean_token_accuracy": 0.08992773070931434,
|
|
"num_tokens": 965565.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.159026861190796,
|
|
"epoch": 0.4437720625315179,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.78,
|
|
"mean_token_accuracy": 0.08791229128837585,
|
|
"num_tokens": 977101.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.1419289112091064,
|
|
"epoch": 0.44881492687846697,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.8314,
|
|
"mean_token_accuracy": 0.08858354166150093,
|
|
"num_tokens": 987445.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.118024682998657,
|
|
"epoch": 0.45385779122541603,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.7034,
|
|
"mean_token_accuracy": 0.09504426047205924,
|
|
"num_tokens": 999379.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.998180723190307,
|
|
"epoch": 0.4589006555723651,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.6412,
|
|
"mean_token_accuracy": 0.09876157343387604,
|
|
"num_tokens": 1010744.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.0912974834442135,
|
|
"epoch": 0.46394351991931415,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.7395,
|
|
"mean_token_accuracy": 0.0892062947154045,
|
|
"num_tokens": 1021808.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.138762521743774,
|
|
"epoch": 0.4689863842662632,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.787,
|
|
"mean_token_accuracy": 0.0881411112844944,
|
|
"num_tokens": 1033694.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.05295991897583,
|
|
"epoch": 0.47402924861321233,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.7531,
|
|
"mean_token_accuracy": 0.09112738594412803,
|
|
"num_tokens": 1044690.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.110706329345703,
|
|
"epoch": 0.4790721129601614,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.7543,
|
|
"mean_token_accuracy": 0.0913363330066204,
|
|
"num_tokens": 1055557.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.020756387710572,
|
|
"epoch": 0.48411497730711045,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.6926,
|
|
"mean_token_accuracy": 0.09522736445069313,
|
|
"num_tokens": 1067304.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.031440687179566,
|
|
"epoch": 0.4891578416540595,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.7153,
|
|
"mean_token_accuracy": 0.09247291311621667,
|
|
"num_tokens": 1078955.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.019560098648071,
|
|
"epoch": 0.49420070600100857,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.7608,
|
|
"mean_token_accuracy": 0.09506335482001305,
|
|
"num_tokens": 1091405.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.067246627807617,
|
|
"epoch": 0.49924357034795763,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.7344,
|
|
"mean_token_accuracy": 0.09412561133503913,
|
|
"num_tokens": 1103227.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.013670969009399,
|
|
"epoch": 0.5042864346949067,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.6734,
|
|
"mean_token_accuracy": 0.09890775382518768,
|
|
"num_tokens": 1114485.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5042864346949067,
|
|
"eval_entropy": 6.978830953894074,
|
|
"eval_loss": 6.783904552459717,
|
|
"eval_mean_token_accuracy": 0.09532982510995622,
|
|
"eval_num_tokens": 1114485.0,
|
|
"eval_runtime": 2.7002,
|
|
"eval_samples_per_second": 1162.13,
|
|
"eval_steps_per_second": 145.544,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.006657695770263,
|
|
"epoch": 0.5093292990418558,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.6997,
|
|
"mean_token_accuracy": 0.0941694438457489,
|
|
"num_tokens": 1126789.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.028399562835693,
|
|
"epoch": 0.5143721633888049,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.6884,
|
|
"mean_token_accuracy": 0.09093117266893387,
|
|
"num_tokens": 1138192.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.900946187973022,
|
|
"epoch": 0.5194150277357539,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.6217,
|
|
"mean_token_accuracy": 0.09346246346831322,
|
|
"num_tokens": 1148526.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.066805505752564,
|
|
"epoch": 0.524457892082703,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.7189,
|
|
"mean_token_accuracy": 0.08981634229421616,
|
|
"num_tokens": 1159194.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.8563981533050535,
|
|
"epoch": 0.529500756429652,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.6209,
|
|
"mean_token_accuracy": 0.09972566738724709,
|
|
"num_tokens": 1170601.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.004376411437988,
|
|
"epoch": 0.5345436207766011,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.6214,
|
|
"mean_token_accuracy": 0.09583464860916138,
|
|
"num_tokens": 1182441.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.822463846206665,
|
|
"epoch": 0.5395864851235502,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.535,
|
|
"mean_token_accuracy": 0.10155294314026833,
|
|
"num_tokens": 1192530.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.8695042610168455,
|
|
"epoch": 0.5446293494704992,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.6516,
|
|
"mean_token_accuracy": 0.09302628561854362,
|
|
"num_tokens": 1203831.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.884985733032226,
|
|
"epoch": 0.5496722138174484,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.5849,
|
|
"mean_token_accuracy": 0.09827456548810005,
|
|
"num_tokens": 1214164.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.899946403503418,
|
|
"epoch": 0.5547150781643974,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.6103,
|
|
"mean_token_accuracy": 0.09503386095166207,
|
|
"num_tokens": 1226932.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.018468141555786,
|
|
"epoch": 0.5597579425113465,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.6799,
|
|
"mean_token_accuracy": 0.09099587425589561,
|
|
"num_tokens": 1237212.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.843923902511596,
|
|
"epoch": 0.5648008068582955,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.5802,
|
|
"mean_token_accuracy": 0.095969757437706,
|
|
"num_tokens": 1249340.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.853866147994995,
|
|
"epoch": 0.5698436712052446,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.5839,
|
|
"mean_token_accuracy": 0.0990279607474804,
|
|
"num_tokens": 1260778.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.846533918380738,
|
|
"epoch": 0.5748865355521936,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.5885,
|
|
"mean_token_accuracy": 0.0959065780043602,
|
|
"num_tokens": 1272216.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.885388469696045,
|
|
"epoch": 0.5799293998991427,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.6706,
|
|
"mean_token_accuracy": 0.09878646060824395,
|
|
"num_tokens": 1283558.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.949707126617431,
|
|
"epoch": 0.5849722642460918,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.729,
|
|
"mean_token_accuracy": 0.09222633503377438,
|
|
"num_tokens": 1294638.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.937385559082031,
|
|
"epoch": 0.5900151285930408,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.6493,
|
|
"mean_token_accuracy": 0.09726304411888123,
|
|
"num_tokens": 1304697.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.841894626617432,
|
|
"epoch": 0.59505799293999,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.6087,
|
|
"mean_token_accuracy": 0.09355146735906601,
|
|
"num_tokens": 1316618.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.907530450820923,
|
|
"epoch": 0.600100857286939,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.7174,
|
|
"mean_token_accuracy": 0.08738479986786843,
|
|
"num_tokens": 1328134.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.918644523620605,
|
|
"epoch": 0.6051437216338881,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.6175,
|
|
"mean_token_accuracy": 0.09422577098011971,
|
|
"num_tokens": 1339101.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.793248319625855,
|
|
"epoch": 0.6101865859808371,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.4922,
|
|
"mean_token_accuracy": 0.09907795786857605,
|
|
"num_tokens": 1349181.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.85523247718811,
|
|
"epoch": 0.6152294503277862,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.6357,
|
|
"mean_token_accuracy": 0.10059619322419167,
|
|
"num_tokens": 1361443.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.753997278213501,
|
|
"epoch": 0.6202723146747352,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.5688,
|
|
"mean_token_accuracy": 0.10217763558030128,
|
|
"num_tokens": 1372807.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.8447350025177,
|
|
"epoch": 0.6253151790216843,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.6514,
|
|
"mean_token_accuracy": 0.09630123376846314,
|
|
"num_tokens": 1385509.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.80165662765503,
|
|
"epoch": 0.6303580433686334,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.5447,
|
|
"mean_token_accuracy": 0.09622347950935364,
|
|
"num_tokens": 1396035.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.750753402709961,
|
|
"epoch": 0.6354009077155824,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.5012,
|
|
"mean_token_accuracy": 0.10519321039319038,
|
|
"num_tokens": 1406839.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.73918046951294,
|
|
"epoch": 0.6404437720625316,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.4651,
|
|
"mean_token_accuracy": 0.09995857030153274,
|
|
"num_tokens": 1417708.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.814885568618775,
|
|
"epoch": 0.6454866364094806,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.5956,
|
|
"mean_token_accuracy": 0.09352405592799187,
|
|
"num_tokens": 1428985.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.812655353546143,
|
|
"epoch": 0.6505295007564297,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.6272,
|
|
"mean_token_accuracy": 0.10007955580949783,
|
|
"num_tokens": 1440232.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.6787721633911135,
|
|
"epoch": 0.6555723651033787,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.3639,
|
|
"mean_token_accuracy": 0.10900432094931603,
|
|
"num_tokens": 1451735.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.713140487670898,
|
|
"epoch": 0.6606152294503278,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.5218,
|
|
"mean_token_accuracy": 0.09929425492882729,
|
|
"num_tokens": 1463477.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.6867029666900635,
|
|
"epoch": 0.6656580937972768,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.5508,
|
|
"mean_token_accuracy": 0.09417604357004165,
|
|
"num_tokens": 1473802.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.759989881515503,
|
|
"epoch": 0.6707009581442259,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.5538,
|
|
"mean_token_accuracy": 0.09805640652775764,
|
|
"num_tokens": 1485519.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.729797887802124,
|
|
"epoch": 0.675743822491175,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.5546,
|
|
"mean_token_accuracy": 0.09775246679782867,
|
|
"num_tokens": 1496957.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.685538005828858,
|
|
"epoch": 0.680786686838124,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.5624,
|
|
"mean_token_accuracy": 0.1045648567378521,
|
|
"num_tokens": 1508336.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.67319016456604,
|
|
"epoch": 0.6858295511850732,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.5527,
|
|
"mean_token_accuracy": 0.10019075870513916,
|
|
"num_tokens": 1520016.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.664720964431763,
|
|
"epoch": 0.6908724155320222,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.3887,
|
|
"mean_token_accuracy": 0.10588083937764167,
|
|
"num_tokens": 1531877.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.550206899642944,
|
|
"epoch": 0.6959152798789713,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.3834,
|
|
"mean_token_accuracy": 0.1071600653231144,
|
|
"num_tokens": 1541553.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.740174055099487,
|
|
"epoch": 0.7009581442259203,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.5491,
|
|
"mean_token_accuracy": 0.10118654146790504,
|
|
"num_tokens": 1552491.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.554775905609131,
|
|
"epoch": 0.7060010085728694,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.4546,
|
|
"mean_token_accuracy": 0.10349184796214103,
|
|
"num_tokens": 1563695.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.716359710693359,
|
|
"epoch": 0.7110438729198184,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.4772,
|
|
"mean_token_accuracy": 0.10427189692854881,
|
|
"num_tokens": 1573467.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.579839134216309,
|
|
"epoch": 0.7160867372667675,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.4774,
|
|
"mean_token_accuracy": 0.10680482760071755,
|
|
"num_tokens": 1584367.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.704503059387207,
|
|
"epoch": 0.7211296016137166,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.4049,
|
|
"mean_token_accuracy": 0.11087890490889549,
|
|
"num_tokens": 1595301.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.5911847114562985,
|
|
"epoch": 0.7261724659606656,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4628,
|
|
"mean_token_accuracy": 0.09945516511797906,
|
|
"num_tokens": 1605916.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.6797267436981205,
|
|
"epoch": 0.7312153303076148,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.4645,
|
|
"mean_token_accuracy": 0.10250937268137932,
|
|
"num_tokens": 1616455.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.535722160339356,
|
|
"epoch": 0.7362581946545638,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.4454,
|
|
"mean_token_accuracy": 0.1064433366060257,
|
|
"num_tokens": 1627396.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.790402030944824,
|
|
"epoch": 0.7413010590015129,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.5785,
|
|
"mean_token_accuracy": 0.09991721957921981,
|
|
"num_tokens": 1639073.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.5861907482147215,
|
|
"epoch": 0.7463439233484619,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.4448,
|
|
"mean_token_accuracy": 0.10222182348370552,
|
|
"num_tokens": 1649787.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.561935520172119,
|
|
"epoch": 0.751386787695411,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.3867,
|
|
"mean_token_accuracy": 0.10436921566724777,
|
|
"num_tokens": 1658883.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.535987997055054,
|
|
"epoch": 0.75642965204236,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.3549,
|
|
"mean_token_accuracy": 0.1072971761226654,
|
|
"num_tokens": 1669099.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.504691028594971,
|
|
"epoch": 0.7614725163893091,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.373,
|
|
"mean_token_accuracy": 0.10971354767680168,
|
|
"num_tokens": 1680227.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.545934200286865,
|
|
"epoch": 0.7665153807362582,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.3708,
|
|
"mean_token_accuracy": 0.10627648830413819,
|
|
"num_tokens": 1689901.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.492063522338867,
|
|
"epoch": 0.7715582450832073,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.4469,
|
|
"mean_token_accuracy": 0.10555314868688584,
|
|
"num_tokens": 1700498.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.565538930892944,
|
|
"epoch": 0.7766011094301564,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.3697,
|
|
"mean_token_accuracy": 0.10695412680506707,
|
|
"num_tokens": 1710715.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.589845371246338,
|
|
"epoch": 0.7816439737771054,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.4913,
|
|
"mean_token_accuracy": 0.1047224260866642,
|
|
"num_tokens": 1723858.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.568187999725342,
|
|
"epoch": 0.7866868381240545,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.4495,
|
|
"mean_token_accuracy": 0.09719336107373237,
|
|
"num_tokens": 1734678.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.497055625915527,
|
|
"epoch": 0.7917297024710035,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.3665,
|
|
"mean_token_accuracy": 0.1123119130730629,
|
|
"num_tokens": 1743869.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.632794427871704,
|
|
"epoch": 0.7967725668179526,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.4189,
|
|
"mean_token_accuracy": 0.1059940479695797,
|
|
"num_tokens": 1755266.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.40299015045166,
|
|
"epoch": 0.8018154311649016,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.3884,
|
|
"mean_token_accuracy": 0.10287827029824256,
|
|
"num_tokens": 1766089.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.549517488479614,
|
|
"epoch": 0.8068582955118507,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.3782,
|
|
"mean_token_accuracy": 0.10602757111191749,
|
|
"num_tokens": 1776057.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.550864315032959,
|
|
"epoch": 0.8119011598587998,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.4463,
|
|
"mean_token_accuracy": 0.10098104774951935,
|
|
"num_tokens": 1786729.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.510504293441772,
|
|
"epoch": 0.8169440242057489,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.3717,
|
|
"mean_token_accuracy": 0.10247773975133896,
|
|
"num_tokens": 1797708.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.48493595123291,
|
|
"epoch": 0.821986888552698,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.4006,
|
|
"mean_token_accuracy": 0.11212195381522179,
|
|
"num_tokens": 1807326.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.447803640365601,
|
|
"epoch": 0.827029752899647,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.3169,
|
|
"mean_token_accuracy": 0.11085391044616699,
|
|
"num_tokens": 1817266.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.538650226593018,
|
|
"epoch": 0.8320726172465961,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.3561,
|
|
"mean_token_accuracy": 0.11818044185638428,
|
|
"num_tokens": 1827840.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.338764762878418,
|
|
"epoch": 0.8371154815935451,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.3387,
|
|
"mean_token_accuracy": 0.11126410290598869,
|
|
"num_tokens": 1838158.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.599035263061523,
|
|
"epoch": 0.8421583459404942,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.4225,
|
|
"mean_token_accuracy": 0.10363714173436164,
|
|
"num_tokens": 1848317.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.398059988021851,
|
|
"epoch": 0.8472012102874432,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.3394,
|
|
"mean_token_accuracy": 0.10549739152193069,
|
|
"num_tokens": 1858751.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.4699845790863035,
|
|
"epoch": 0.8522440746343923,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.365,
|
|
"mean_token_accuracy": 0.1064521923661232,
|
|
"num_tokens": 1870739.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.610909032821655,
|
|
"epoch": 0.8572869389813415,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.4852,
|
|
"mean_token_accuracy": 0.10624399855732918,
|
|
"num_tokens": 1880995.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.465567302703858,
|
|
"epoch": 0.8623298033282905,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.4162,
|
|
"mean_token_accuracy": 0.10567688420414925,
|
|
"num_tokens": 1892633.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.5905781269073485,
|
|
"epoch": 0.8673726676752396,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.5011,
|
|
"mean_token_accuracy": 0.09975023046135903,
|
|
"num_tokens": 1904802.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.505033540725708,
|
|
"epoch": 0.8724155320221886,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.492,
|
|
"mean_token_accuracy": 0.10669593513011932,
|
|
"num_tokens": 1915755.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.387980556488037,
|
|
"epoch": 0.8774583963691377,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.2328,
|
|
"mean_token_accuracy": 0.10829911306500435,
|
|
"num_tokens": 1926098.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.521399021148682,
|
|
"epoch": 0.8825012607160867,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.4569,
|
|
"mean_token_accuracy": 0.10727727934718131,
|
|
"num_tokens": 1936607.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.403067350387573,
|
|
"epoch": 0.8875441250630358,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.3421,
|
|
"mean_token_accuracy": 0.10624040365219116,
|
|
"num_tokens": 1948596.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.43417534828186,
|
|
"epoch": 0.8925869894099848,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.4263,
|
|
"mean_token_accuracy": 0.11106917932629586,
|
|
"num_tokens": 1960322.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.524694728851318,
|
|
"epoch": 0.8976298537569339,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.3527,
|
|
"mean_token_accuracy": 0.11080583110451699,
|
|
"num_tokens": 1971052.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.44118709564209,
|
|
"epoch": 0.902672718103883,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.3485,
|
|
"mean_token_accuracy": 0.10465901717543602,
|
|
"num_tokens": 1980917.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.377262210845947,
|
|
"epoch": 0.9077155824508321,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.3261,
|
|
"mean_token_accuracy": 0.11112496107816697,
|
|
"num_tokens": 1991035.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.4430749893188475,
|
|
"epoch": 0.9127584467977812,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.3419,
|
|
"mean_token_accuracy": 0.10812708809971809,
|
|
"num_tokens": 2000587.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.400346374511718,
|
|
"epoch": 0.9178013111447302,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.3057,
|
|
"mean_token_accuracy": 0.11217240318655967,
|
|
"num_tokens": 2011087.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.43453688621521,
|
|
"epoch": 0.9228441754916793,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.3459,
|
|
"mean_token_accuracy": 0.10732184424996376,
|
|
"num_tokens": 2023472.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.432980585098266,
|
|
"epoch": 0.9278870398386283,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.3346,
|
|
"mean_token_accuracy": 0.1071820467710495,
|
|
"num_tokens": 2033783.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.326636266708374,
|
|
"epoch": 0.9329299041855774,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.3581,
|
|
"mean_token_accuracy": 0.11071096733212471,
|
|
"num_tokens": 2045882.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.479465627670288,
|
|
"epoch": 0.9379727685325264,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.3877,
|
|
"mean_token_accuracy": 0.10236008018255234,
|
|
"num_tokens": 2056549.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.462947130203247,
|
|
"epoch": 0.9430156328794755,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.3814,
|
|
"mean_token_accuracy": 0.11182121336460113,
|
|
"num_tokens": 2067540.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.290677881240844,
|
|
"epoch": 0.9480584972264247,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.276,
|
|
"mean_token_accuracy": 0.10857773795723916,
|
|
"num_tokens": 2077386.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.355180120468139,
|
|
"epoch": 0.9531013615733737,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.2417,
|
|
"mean_token_accuracy": 0.11178631335496902,
|
|
"num_tokens": 2087777.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.281049156188965,
|
|
"epoch": 0.9581442259203228,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.2404,
|
|
"mean_token_accuracy": 0.11323176622390747,
|
|
"num_tokens": 2099071.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.204891157150269,
|
|
"epoch": 0.9631870902672718,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.3123,
|
|
"mean_token_accuracy": 0.1093476541340351,
|
|
"num_tokens": 2109168.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.614584875106812,
|
|
"epoch": 0.9682299546142209,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.4215,
|
|
"mean_token_accuracy": 0.10730769336223603,
|
|
"num_tokens": 2120696.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.28635835647583,
|
|
"epoch": 0.9732728189611699,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.37,
|
|
"mean_token_accuracy": 0.10793575569987297,
|
|
"num_tokens": 2132427.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.3901630401611325,
|
|
"epoch": 0.978315683308119,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.3436,
|
|
"mean_token_accuracy": 0.1093747541308403,
|
|
"num_tokens": 2143120.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.467719650268554,
|
|
"epoch": 0.983358547655068,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.3473,
|
|
"mean_token_accuracy": 0.10565910637378692,
|
|
"num_tokens": 2154154.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.314101648330689,
|
|
"epoch": 0.9884014120020171,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.4482,
|
|
"mean_token_accuracy": 0.10238303095102311,
|
|
"num_tokens": 2167000.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.403873634338379,
|
|
"epoch": 0.9934442763489663,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.3777,
|
|
"mean_token_accuracy": 0.10910984128713608,
|
|
"num_tokens": 2178100.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.348555374145508,
|
|
"epoch": 0.9984871406959153,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3173,
|
|
"mean_token_accuracy": 0.10381858721375466,
|
|
"num_tokens": 2189063.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.4779570367601185,
|
|
"epoch": 1.0030257186081695,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.2356,
|
|
"mean_token_accuracy": 0.11199044684569041,
|
|
"num_tokens": 2198302.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.333552265167237,
|
|
"epoch": 1.0080685829551186,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.1906,
|
|
"mean_token_accuracy": 0.1150072768330574,
|
|
"num_tokens": 2209648.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0080685829551186,
|
|
"eval_entropy": 6.4261680559347605,
|
|
"eval_loss": 6.417810440063477,
|
|
"eval_mean_token_accuracy": 0.10917834623813932,
|
|
"eval_num_tokens": 2209648.0,
|
|
"eval_runtime": 2.6477,
|
|
"eval_samples_per_second": 1185.162,
|
|
"eval_steps_per_second": 148.429,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.252728891372681,
|
|
"epoch": 1.0131114473020675,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999997762223182,
|
|
"loss": 6.1308,
|
|
"mean_token_accuracy": 0.11400617361068725,
|
|
"num_tokens": 2220806.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.257132434844971,
|
|
"epoch": 1.0181543116490166,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999988671262489,
|
|
"loss": 6.1254,
|
|
"mean_token_accuracy": 0.11259887367486954,
|
|
"num_tokens": 2231806.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.295977258682251,
|
|
"epoch": 1.0231971759959657,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999972587285105,
|
|
"loss": 6.1821,
|
|
"mean_token_accuracy": 0.11922840997576714,
|
|
"num_tokens": 2243835.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.117448616027832,
|
|
"epoch": 1.0282400403429148,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999949510341015,
|
|
"loss": 6.0891,
|
|
"mean_token_accuracy": 0.11738667264580727,
|
|
"num_tokens": 2254722.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.269881677627564,
|
|
"epoch": 1.033282904689864,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999919440501947,
|
|
"loss": 6.0294,
|
|
"mean_token_accuracy": 0.12477270513772964,
|
|
"num_tokens": 2264806.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.358243036270141,
|
|
"epoch": 1.0383257690368128,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999882377861355,
|
|
"loss": 6.2479,
|
|
"mean_token_accuracy": 0.11410975456237793,
|
|
"num_tokens": 2277037.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.277375936508179,
|
|
"epoch": 1.043368633383762,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999838322534433,
|
|
"loss": 6.1785,
|
|
"mean_token_accuracy": 0.1084089681506157,
|
|
"num_tokens": 2289829.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.295744276046753,
|
|
"epoch": 1.048411497730711,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999787274658107,
|
|
"loss": 6.0711,
|
|
"mean_token_accuracy": 0.11793513298034668,
|
|
"num_tokens": 2300340.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.199061441421509,
|
|
"epoch": 1.0534543620776602,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999729234391032,
|
|
"loss": 6.1059,
|
|
"mean_token_accuracy": 0.12160981521010399,
|
|
"num_tokens": 2310009.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.198751163482666,
|
|
"epoch": 1.058497226424609,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00049996642019136,
|
|
"loss": 6.1428,
|
|
"mean_token_accuracy": 0.11531612128019333,
|
|
"num_tokens": 2319290.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.297437572479248,
|
|
"epoch": 1.0635400907715582,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999592177427933,
|
|
"loss": 6.0857,
|
|
"mean_token_accuracy": 0.12294981405138969,
|
|
"num_tokens": 2329170.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.1483875751495365,
|
|
"epoch": 1.0685829551185073,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999513161157885,
|
|
"loss": 6.105,
|
|
"mean_token_accuracy": 0.11440970972180367,
|
|
"num_tokens": 2339617.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.260158061981201,
|
|
"epoch": 1.0736258194654564,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999427153349041,
|
|
"loss": 6.0714,
|
|
"mean_token_accuracy": 0.12414619326591492,
|
|
"num_tokens": 2349548.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.0723645210266115,
|
|
"epoch": 1.0786686838124055,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999334154268714,
|
|
"loss": 6.0752,
|
|
"mean_token_accuracy": 0.11783618927001953,
|
|
"num_tokens": 2359116.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.276753377914429,
|
|
"epoch": 1.0837115481593544,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999234164205949,
|
|
"loss": 6.1327,
|
|
"mean_token_accuracy": 0.12174399495124817,
|
|
"num_tokens": 2369412.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.1624593257904055,
|
|
"epoch": 1.0887544125063036,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999127183471517,
|
|
"loss": 6.2097,
|
|
"mean_token_accuracy": 0.11544428989291192,
|
|
"num_tokens": 2381276.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.245842838287354,
|
|
"epoch": 1.0937972768532527,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999013212397915,
|
|
"loss": 6.1076,
|
|
"mean_token_accuracy": 0.11853586509823799,
|
|
"num_tokens": 2392135.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.093086051940918,
|
|
"epoch": 1.0988401412002018,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004998892251339368,
|
|
"loss": 6.0149,
|
|
"mean_token_accuracy": 0.1269606776535511,
|
|
"num_tokens": 2402957.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.243423843383789,
|
|
"epoch": 1.1038830055471507,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998764300671827,
|
|
"loss": 6.1579,
|
|
"mean_token_accuracy": 0.12277726605534553,
|
|
"num_tokens": 2413386.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.157353162765503,
|
|
"epoch": 1.1089258698940998,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998629360792965,
|
|
"loss": 6.1449,
|
|
"mean_token_accuracy": 0.11866232082247734,
|
|
"num_tokens": 2424441.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.189147281646728,
|
|
"epoch": 1.113968734241049,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998487432122179,
|
|
"loss": 6.1766,
|
|
"mean_token_accuracy": 0.11409586071968078,
|
|
"num_tokens": 2436144.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.2349693298339846,
|
|
"epoch": 1.119011598587998,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004998338515100585,
|
|
"loss": 6.0813,
|
|
"mean_token_accuracy": 0.12133537903428078,
|
|
"num_tokens": 2447393.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.068421173095703,
|
|
"epoch": 1.1240544629349472,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998182610191022,
|
|
"loss": 6.0428,
|
|
"mean_token_accuracy": 0.12405308485031127,
|
|
"num_tokens": 2458632.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.2577080726623535,
|
|
"epoch": 1.129097327281896,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998019717878047,
|
|
"loss": 6.1288,
|
|
"mean_token_accuracy": 0.11385903730988503,
|
|
"num_tokens": 2469331.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.227204322814941,
|
|
"epoch": 1.1341401916288452,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000499784983866793,
|
|
"loss": 6.1671,
|
|
"mean_token_accuracy": 0.11437357068061829,
|
|
"num_tokens": 2479501.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.0807130336761475,
|
|
"epoch": 1.1391830559757943,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997672973088661,
|
|
"loss": 6.0287,
|
|
"mean_token_accuracy": 0.12107328623533249,
|
|
"num_tokens": 2490087.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.214322185516357,
|
|
"epoch": 1.1442259203227434,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004997489121689944,
|
|
"loss": 6.0967,
|
|
"mean_token_accuracy": 0.11605892032384872,
|
|
"num_tokens": 2501130.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.246376657485962,
|
|
"epoch": 1.1492687846696923,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997298285043189,
|
|
"loss": 6.2169,
|
|
"mean_token_accuracy": 0.11053206399083138,
|
|
"num_tokens": 2513999.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.166038084030151,
|
|
"epoch": 1.1543116490166414,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997100463741527,
|
|
"loss": 6.1287,
|
|
"mean_token_accuracy": 0.11388898193836212,
|
|
"num_tokens": 2525579.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.167115879058838,
|
|
"epoch": 1.1593545133635905,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996895658399785,
|
|
"loss": 6.0949,
|
|
"mean_token_accuracy": 0.12076913714408874,
|
|
"num_tokens": 2536091.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.126659774780274,
|
|
"epoch": 1.1643973777105396,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004996683869654507,
|
|
"loss": 6.1302,
|
|
"mean_token_accuracy": 0.12046247944235802,
|
|
"num_tokens": 2546073.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.1831365585327145,
|
|
"epoch": 1.1694402420574885,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996465098163934,
|
|
"loss": 6.0907,
|
|
"mean_token_accuracy": 0.11801456063985824,
|
|
"num_tokens": 2556626.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.1779944896698,
|
|
"epoch": 1.1744831064044376,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004996239344608015,
|
|
"loss": 6.1206,
|
|
"mean_token_accuracy": 0.12329623401165009,
|
|
"num_tokens": 2567304.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.129946279525757,
|
|
"epoch": 1.1795259707513868,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004996006609688398,
|
|
"loss": 5.9896,
|
|
"mean_token_accuracy": 0.1163274422287941,
|
|
"num_tokens": 2577242.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.059417057037353,
|
|
"epoch": 1.1845688350983359,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004995766894128426,
|
|
"loss": 6.0956,
|
|
"mean_token_accuracy": 0.11244072541594505,
|
|
"num_tokens": 2588330.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.160956144332886,
|
|
"epoch": 1.189611699445285,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995520198673143,
|
|
"loss": 6.0936,
|
|
"mean_token_accuracy": 0.12082465887069702,
|
|
"num_tokens": 2598471.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.21316089630127,
|
|
"epoch": 1.194654563792234,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004995266524089283,
|
|
"loss": 6.1817,
|
|
"mean_token_accuracy": 0.11847011670470238,
|
|
"num_tokens": 2610238.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.094186544418335,
|
|
"epoch": 1.199697428139183,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004995005871165271,
|
|
"loss": 6.0416,
|
|
"mean_token_accuracy": 0.1227484904229641,
|
|
"num_tokens": 2619870.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.185560274124145,
|
|
"epoch": 1.2047402924861321,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004994738240711225,
|
|
"loss": 6.1908,
|
|
"mean_token_accuracy": 0.11310086920857429,
|
|
"num_tokens": 2631541.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.106654500961303,
|
|
"epoch": 1.2097831568330812,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004994463633558945,
|
|
"loss": 6.0229,
|
|
"mean_token_accuracy": 0.12247754111886025,
|
|
"num_tokens": 2641896.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.082180595397949,
|
|
"epoch": 1.2148260211800301,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004994182050561917,
|
|
"loss": 6.1038,
|
|
"mean_token_accuracy": 0.12054873630404472,
|
|
"num_tokens": 2654208.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.143400192260742,
|
|
"epoch": 1.2198688855269793,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004993893492595306,
|
|
"loss": 6.0785,
|
|
"mean_token_accuracy": 0.12056932002305984,
|
|
"num_tokens": 2665115.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.13805513381958,
|
|
"epoch": 1.2249117498739284,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004993597960555959,
|
|
"loss": 6.1659,
|
|
"mean_token_accuracy": 0.11709513887763023,
|
|
"num_tokens": 2676093.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.216587829589844,
|
|
"epoch": 1.2299546142208775,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004993295455362395,
|
|
"loss": 6.1297,
|
|
"mean_token_accuracy": 0.12260301932692527,
|
|
"num_tokens": 2687866.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.135248184204102,
|
|
"epoch": 1.2349974785678266,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004992985977954808,
|
|
"loss": 6.1339,
|
|
"mean_token_accuracy": 0.11458160728216171,
|
|
"num_tokens": 2699814.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.165663337707519,
|
|
"epoch": 1.2400403429147757,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499266952929506,
|
|
"loss": 6.0974,
|
|
"mean_token_accuracy": 0.12700220122933387,
|
|
"num_tokens": 2710384.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.150148534774781,
|
|
"epoch": 1.2450832072617246,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004992346110366682,
|
|
"loss": 6.0655,
|
|
"mean_token_accuracy": 0.11997778415679931,
|
|
"num_tokens": 2721442.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.090733575820923,
|
|
"epoch": 1.2501260716086737,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004992015722174865,
|
|
"loss": 6.0888,
|
|
"mean_token_accuracy": 0.11658885851502418,
|
|
"num_tokens": 2733605.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.171622323989868,
|
|
"epoch": 1.2551689359556228,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004991678365746466,
|
|
"loss": 6.128,
|
|
"mean_token_accuracy": 0.12494012638926506,
|
|
"num_tokens": 2744853.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.0140642642974855,
|
|
"epoch": 1.2602118003025717,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004991334042129996,
|
|
"loss": 6.023,
|
|
"mean_token_accuracy": 0.12799779549241067,
|
|
"num_tokens": 2754477.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.170590877532959,
|
|
"epoch": 1.2652546646495209,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004990982752395618,
|
|
"loss": 6.0337,
|
|
"mean_token_accuracy": 0.12020832374691963,
|
|
"num_tokens": 2765895.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.077351331710815,
|
|
"epoch": 1.27029752899647,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004990624497635152,
|
|
"loss": 6.0866,
|
|
"mean_token_accuracy": 0.11977267414331436,
|
|
"num_tokens": 2776897.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.209859228134155,
|
|
"epoch": 1.275340393343419,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000499025927896206,
|
|
"loss": 6.1038,
|
|
"mean_token_accuracy": 0.12009399533271789,
|
|
"num_tokens": 2788666.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.031207752227783,
|
|
"epoch": 1.2803832576903682,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004989887097511451,
|
|
"loss": 5.9986,
|
|
"mean_token_accuracy": 0.12084411457180977,
|
|
"num_tokens": 2799167.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.0679034233093265,
|
|
"epoch": 1.2854261220373173,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004989507954440074,
|
|
"loss": 6.0037,
|
|
"mean_token_accuracy": 0.11892658248543739,
|
|
"num_tokens": 2809774.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.099821901321411,
|
|
"epoch": 1.2904689863842662,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004989121850926313,
|
|
"loss": 6.1233,
|
|
"mean_token_accuracy": 0.11551008000969887,
|
|
"num_tokens": 2820515.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.086860656738281,
|
|
"epoch": 1.2955118507312153,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004988728788170188,
|
|
"loss": 6.0611,
|
|
"mean_token_accuracy": 0.12110966295003892,
|
|
"num_tokens": 2832163.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.091725492477417,
|
|
"epoch": 1.3005547150781644,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004988328767393346,
|
|
"loss": 6.0574,
|
|
"mean_token_accuracy": 0.12182244136929513,
|
|
"num_tokens": 2843717.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 5.981151485443116,
|
|
"epoch": 1.3055975794251133,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004987921789839061,
|
|
"loss": 6.0158,
|
|
"mean_token_accuracy": 0.12019825726747513,
|
|
"num_tokens": 2854489.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.1015993595123295,
|
|
"epoch": 1.3106404437720625,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004987507856772229,
|
|
"loss": 6.0627,
|
|
"mean_token_accuracy": 0.1227644480764866,
|
|
"num_tokens": 2865317.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.058894157409668,
|
|
"epoch": 1.3156833081190116,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004987086969479364,
|
|
"loss": 5.9809,
|
|
"mean_token_accuracy": 0.12572994455695152,
|
|
"num_tokens": 2877466.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.00954852104187,
|
|
"epoch": 1.3207261724659607,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004986659129268592,
|
|
"loss": 6.0129,
|
|
"mean_token_accuracy": 0.12269562259316444,
|
|
"num_tokens": 2887698.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 5.980123376846313,
|
|
"epoch": 1.3257690368129098,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004986224337469652,
|
|
"loss": 5.9609,
|
|
"mean_token_accuracy": 0.12754920348525048,
|
|
"num_tokens": 2899339.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.056573486328125,
|
|
"epoch": 1.330811901159859,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004985782595433886,
|
|
"loss": 6.0081,
|
|
"mean_token_accuracy": 0.12356857880949974,
|
|
"num_tokens": 2908340.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.078930139541626,
|
|
"epoch": 1.3358547655068078,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004985333904534237,
|
|
"loss": 6.1003,
|
|
"mean_token_accuracy": 0.12403591498732566,
|
|
"num_tokens": 2920432.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.12197904586792,
|
|
"epoch": 1.340897629853757,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000498487826616525,
|
|
"loss": 6.055,
|
|
"mean_token_accuracy": 0.12046747058629989,
|
|
"num_tokens": 2930385.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.096244239807129,
|
|
"epoch": 1.345940494200706,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004984415681743056,
|
|
"loss": 5.9999,
|
|
"mean_token_accuracy": 0.12470603808760643,
|
|
"num_tokens": 2940760.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.053582334518433,
|
|
"epoch": 1.350983358547655,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004983946152705382,
|
|
"loss": 6.0303,
|
|
"mean_token_accuracy": 0.1265427976846695,
|
|
"num_tokens": 2952441.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.091333484649658,
|
|
"epoch": 1.356026222894604,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004983469680511531,
|
|
"loss": 6.0466,
|
|
"mean_token_accuracy": 0.12300621345639229,
|
|
"num_tokens": 2962766.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 5.971969509124756,
|
|
"epoch": 1.3610690872415532,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004982986266642393,
|
|
"loss": 5.9324,
|
|
"mean_token_accuracy": 0.12989450618624687,
|
|
"num_tokens": 2974239.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.044353055953979,
|
|
"epoch": 1.3661119515885023,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004982495912600426,
|
|
"loss": 5.9819,
|
|
"mean_token_accuracy": 0.12527156621217728,
|
|
"num_tokens": 2984497.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.031408786773682,
|
|
"epoch": 1.3711548159354514,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004981998619909664,
|
|
"loss": 5.9519,
|
|
"mean_token_accuracy": 0.13078300505876542,
|
|
"num_tokens": 2995367.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.021449947357178,
|
|
"epoch": 1.3761976802824005,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004981494390115704,
|
|
"loss": 5.9832,
|
|
"mean_token_accuracy": 0.1252884514629841,
|
|
"num_tokens": 3007687.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.037176036834717,
|
|
"epoch": 1.3812405446293494,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004980983224785704,
|
|
"loss": 5.9876,
|
|
"mean_token_accuracy": 0.12274198979139328,
|
|
"num_tokens": 3017505.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 5.987515687942505,
|
|
"epoch": 1.3862834089762985,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004980465125508376,
|
|
"loss": 5.9261,
|
|
"mean_token_accuracy": 0.12372706532478332,
|
|
"num_tokens": 3028271.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.0109961986541744,
|
|
"epoch": 1.3913262733232477,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004979940093893986,
|
|
"loss": 5.9846,
|
|
"mean_token_accuracy": 0.12520991489291192,
|
|
"num_tokens": 3040440.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 5.929500961303711,
|
|
"epoch": 1.3963691376701965,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004979408131574345,
|
|
"loss": 5.9556,
|
|
"mean_token_accuracy": 0.12497764006257057,
|
|
"num_tokens": 3050999.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.053280925750732,
|
|
"epoch": 1.4014120020171457,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004978869240202802,
|
|
"loss": 5.9992,
|
|
"mean_token_accuracy": 0.12776673436164857,
|
|
"num_tokens": 3063374.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.028737211227417,
|
|
"epoch": 1.4064548663640948,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004978323421454246,
|
|
"loss": 6.0395,
|
|
"mean_token_accuracy": 0.12454756051301956,
|
|
"num_tokens": 3075464.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.949777460098266,
|
|
"epoch": 1.411497730711044,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004977770677025092,
|
|
"loss": 5.9103,
|
|
"mean_token_accuracy": 0.13049111217260362,
|
|
"num_tokens": 3086398.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 5.946027517318726,
|
|
"epoch": 1.416540595057993,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004977211008633284,
|
|
"loss": 6.0461,
|
|
"mean_token_accuracy": 0.11787689551711082,
|
|
"num_tokens": 3097777.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.0807273387908936,
|
|
"epoch": 1.4215834594049421,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004976644418018285,
|
|
"loss": 6.0026,
|
|
"mean_token_accuracy": 0.13193272352218627,
|
|
"num_tokens": 3108711.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.04375581741333,
|
|
"epoch": 1.426626323751891,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000497607090694107,
|
|
"loss": 6.0231,
|
|
"mean_token_accuracy": 0.12179360166192055,
|
|
"num_tokens": 3119589.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.017640018463135,
|
|
"epoch": 1.4316691880988401,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004975490477184126,
|
|
"loss": 5.9568,
|
|
"mean_token_accuracy": 0.13265052065253258,
|
|
"num_tokens": 3131744.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.093933820724487,
|
|
"epoch": 1.4367120524457893,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000497490313055144,
|
|
"loss": 6.003,
|
|
"mean_token_accuracy": 0.12273363247513772,
|
|
"num_tokens": 3143526.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.946259832382202,
|
|
"epoch": 1.4417549167927382,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004974308868868501,
|
|
"loss": 5.9407,
|
|
"mean_token_accuracy": 0.12551197931170463,
|
|
"num_tokens": 3154561.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.044280624389648,
|
|
"epoch": 1.4467977811396873,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004973707693982287,
|
|
"loss": 6.1069,
|
|
"mean_token_accuracy": 0.1201934814453125,
|
|
"num_tokens": 3166052.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.130475854873657,
|
|
"epoch": 1.4518406454866364,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004973099607761264,
|
|
"loss": 6.0729,
|
|
"mean_token_accuracy": 0.12008198499679565,
|
|
"num_tokens": 3179335.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.9311377048492435,
|
|
"epoch": 1.4568835098335855,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004972484612095378,
|
|
"loss": 5.957,
|
|
"mean_token_accuracy": 0.12554447948932648,
|
|
"num_tokens": 3192129.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.110207796096802,
|
|
"epoch": 1.4619263741805346,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000497186270889605,
|
|
"loss": 6.0633,
|
|
"mean_token_accuracy": 0.12692347541451454,
|
|
"num_tokens": 3202792.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.0356773853302,
|
|
"epoch": 1.4669692385274837,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004971233900096169,
|
|
"loss": 6.0074,
|
|
"mean_token_accuracy": 0.12310663536190987,
|
|
"num_tokens": 3213707.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.061852645874024,
|
|
"epoch": 1.4720121028744326,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004970598187650085,
|
|
"loss": 5.9785,
|
|
"mean_token_accuracy": 0.1272933304309845,
|
|
"num_tokens": 3224531.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 5.997699165344239,
|
|
"epoch": 1.4770549672213817,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000496995557353361,
|
|
"loss": 5.9685,
|
|
"mean_token_accuracy": 0.12390959337353706,
|
|
"num_tokens": 3234656.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.925370979309082,
|
|
"epoch": 1.4820978315683309,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004969306059744004,
|
|
"loss": 5.9988,
|
|
"mean_token_accuracy": 0.12714771330356597,
|
|
"num_tokens": 3246745.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.007743883132934,
|
|
"epoch": 1.4871406959152798,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004968649648299968,
|
|
"loss": 5.9677,
|
|
"mean_token_accuracy": 0.12777481228113174,
|
|
"num_tokens": 3257277.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.022071361541748,
|
|
"epoch": 1.4921835602622289,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004967986341241646,
|
|
"loss": 5.9527,
|
|
"mean_token_accuracy": 0.12892103865742682,
|
|
"num_tokens": 3267704.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.055181169509888,
|
|
"epoch": 1.497226424609178,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004967316140630611,
|
|
"loss": 6.0201,
|
|
"mean_token_accuracy": 0.1264706738293171,
|
|
"num_tokens": 3279772.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.977132177352905,
|
|
"epoch": 1.502269288956127,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004966639048549862,
|
|
"loss": 5.9196,
|
|
"mean_token_accuracy": 0.13101670071482657,
|
|
"num_tokens": 3290536.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.940878009796142,
|
|
"epoch": 1.5073121533030762,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004965955067103816,
|
|
"loss": 6.0031,
|
|
"mean_token_accuracy": 0.13029861822724342,
|
|
"num_tokens": 3302051.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.055119800567627,
|
|
"epoch": 1.5123550176500253,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004965264198418305,
|
|
"loss": 5.9755,
|
|
"mean_token_accuracy": 0.11771604046225548,
|
|
"num_tokens": 3314324.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.5123550176500253,
|
|
"eval_entropy": 5.820931597823709,
|
|
"eval_loss": 6.134761333465576,
|
|
"eval_mean_token_accuracy": 0.125417051083259,
|
|
"eval_num_tokens": 3314324.0,
|
|
"eval_runtime": 2.6525,
|
|
"eval_samples_per_second": 1183.028,
|
|
"eval_steps_per_second": 148.161,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.888900661468506,
|
|
"epoch": 1.5173978819969742,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004964566444640564,
|
|
"loss": 5.9638,
|
|
"mean_token_accuracy": 0.12984286695718766,
|
|
"num_tokens": 3326246.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.980616760253906,
|
|
"epoch": 1.5224407463439233,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004963861807939227,
|
|
"loss": 5.9547,
|
|
"mean_token_accuracy": 0.12878640592098237,
|
|
"num_tokens": 3336308.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.875748062133789,
|
|
"epoch": 1.5274836106908725,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004963150290504321,
|
|
"loss": 5.8941,
|
|
"mean_token_accuracy": 0.1307993695139885,
|
|
"num_tokens": 3347596.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.888290357589722,
|
|
"epoch": 1.5325264750378214,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004962431894547259,
|
|
"loss": 5.8852,
|
|
"mean_token_accuracy": 0.1301721751689911,
|
|
"num_tokens": 3358370.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.044656658172608,
|
|
"epoch": 1.5375693393847705,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004961706622300834,
|
|
"loss": 5.8976,
|
|
"mean_token_accuracy": 0.12725164890289306,
|
|
"num_tokens": 3368630.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 5.95916166305542,
|
|
"epoch": 1.5426122037317196,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004960974476019206,
|
|
"loss": 5.9281,
|
|
"mean_token_accuracy": 0.13028757646679878,
|
|
"num_tokens": 3379952.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.915283250808716,
|
|
"epoch": 1.5476550680786687,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004960235457977905,
|
|
"loss": 5.8544,
|
|
"mean_token_accuracy": 0.1292065441608429,
|
|
"num_tokens": 3390097.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.88209900856018,
|
|
"epoch": 1.5526979324256178,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004959489570473814,
|
|
"loss": 5.8503,
|
|
"mean_token_accuracy": 0.13215067759156227,
|
|
"num_tokens": 3399778.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.941514205932617,
|
|
"epoch": 1.557740796772567,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000495873681582517,
|
|
"loss": 6.007,
|
|
"mean_token_accuracy": 0.12837614566087724,
|
|
"num_tokens": 3410213.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.009626150131226,
|
|
"epoch": 1.5627836611195158,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004957977196371552,
|
|
"loss": 6.0055,
|
|
"mean_token_accuracy": 0.12803423553705215,
|
|
"num_tokens": 3422071.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.056908845901489,
|
|
"epoch": 1.567826525466465,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004957210714473873,
|
|
"loss": 5.9833,
|
|
"mean_token_accuracy": 0.13199174925684928,
|
|
"num_tokens": 3432102.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.918883609771728,
|
|
"epoch": 1.572869389813414,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004956437372514378,
|
|
"loss": 5.9638,
|
|
"mean_token_accuracy": 0.12932863235473632,
|
|
"num_tokens": 3442821.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.923231077194214,
|
|
"epoch": 1.577912254160363,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004955657172896631,
|
|
"loss": 5.9211,
|
|
"mean_token_accuracy": 0.12785825058817862,
|
|
"num_tokens": 3454378.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.945256757736206,
|
|
"epoch": 1.582955118507312,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004954870118045511,
|
|
"loss": 5.941,
|
|
"mean_token_accuracy": 0.13135533407330513,
|
|
"num_tokens": 3463933.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.024400615692139,
|
|
"epoch": 1.5879979828542612,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004954076210407203,
|
|
"loss": 5.8713,
|
|
"mean_token_accuracy": 0.12804725468158723,
|
|
"num_tokens": 3473993.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.913674783706665,
|
|
"epoch": 1.5930408472012103,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004953275452449188,
|
|
"loss": 6.0226,
|
|
"mean_token_accuracy": 0.12443782985210419,
|
|
"num_tokens": 3484882.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.996275997161865,
|
|
"epoch": 1.5980837115481594,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004952467846660243,
|
|
"loss": 5.9353,
|
|
"mean_token_accuracy": 0.12367153093218804,
|
|
"num_tokens": 3496503.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.989222621917724,
|
|
"epoch": 1.6031265758951085,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004951653395550423,
|
|
"loss": 6.016,
|
|
"mean_token_accuracy": 0.1291416108608246,
|
|
"num_tokens": 3508696.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.024582004547119,
|
|
"epoch": 1.6081694402420574,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004950832101651062,
|
|
"loss": 5.9267,
|
|
"mean_token_accuracy": 0.13258333280682563,
|
|
"num_tokens": 3519542.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.871862697601318,
|
|
"epoch": 1.6132123045890066,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004950003967514762,
|
|
"loss": 5.9008,
|
|
"mean_token_accuracy": 0.13136171102523803,
|
|
"num_tokens": 3531576.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.985661792755127,
|
|
"epoch": 1.6182551689359557,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000494916899571538,
|
|
"loss": 5.8534,
|
|
"mean_token_accuracy": 0.13423650190234185,
|
|
"num_tokens": 3542816.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.966505241394043,
|
|
"epoch": 1.6232980332829046,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004948327188848029,
|
|
"loss": 5.9948,
|
|
"mean_token_accuracy": 0.1255824327468872,
|
|
"num_tokens": 3554100.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.077696657180786,
|
|
"epoch": 1.6283408976298537,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004947478549529065,
|
|
"loss": 5.8494,
|
|
"mean_token_accuracy": 0.12402525842189789,
|
|
"num_tokens": 3565048.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.930281019210815,
|
|
"epoch": 1.6333837619768028,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004946623080396077,
|
|
"loss": 5.9554,
|
|
"mean_token_accuracy": 0.12922262623906136,
|
|
"num_tokens": 3575587.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.955634641647339,
|
|
"epoch": 1.638426626323752,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004945760784107884,
|
|
"loss": 6.0183,
|
|
"mean_token_accuracy": 0.12855926603078843,
|
|
"num_tokens": 3587059.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.9010406017303465,
|
|
"epoch": 1.643469490670701,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004944891663344522,
|
|
"loss": 5.8267,
|
|
"mean_token_accuracy": 0.1329905480146408,
|
|
"num_tokens": 3598492.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.891679334640503,
|
|
"epoch": 1.6485123550176501,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004944015720807239,
|
|
"loss": 5.9012,
|
|
"mean_token_accuracy": 0.13142564743757248,
|
|
"num_tokens": 3609513.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.980248069763183,
|
|
"epoch": 1.653555219364599,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004943132959218484,
|
|
"loss": 5.9307,
|
|
"mean_token_accuracy": 0.13032082244753837,
|
|
"num_tokens": 3621327.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.883354759216308,
|
|
"epoch": 1.6585980837115482,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00049422433813219,
|
|
"loss": 5.8355,
|
|
"mean_token_accuracy": 0.1323447398841381,
|
|
"num_tokens": 3632136.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.896560430526733,
|
|
"epoch": 1.6636409480584973,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004941346989882318,
|
|
"loss": 5.8008,
|
|
"mean_token_accuracy": 0.1278294138610363,
|
|
"num_tokens": 3642271.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.7824047088623045,
|
|
"epoch": 1.6686838124054462,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004940443787685739,
|
|
"loss": 5.9415,
|
|
"mean_token_accuracy": 0.12705803215503692,
|
|
"num_tokens": 3652355.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.048564910888672,
|
|
"epoch": 1.6737266767523953,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004939533777539341,
|
|
"loss": 5.9195,
|
|
"mean_token_accuracy": 0.13076957687735558,
|
|
"num_tokens": 3663714.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.867811965942383,
|
|
"epoch": 1.6787695410993444,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004938616962271453,
|
|
"loss": 5.8732,
|
|
"mean_token_accuracy": 0.1254029929637909,
|
|
"num_tokens": 3674675.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.935502529144287,
|
|
"epoch": 1.6838124054462935,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004937693344731561,
|
|
"loss": 5.9575,
|
|
"mean_token_accuracy": 0.1326349377632141,
|
|
"num_tokens": 3686403.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.9425211429595945,
|
|
"epoch": 1.6888552697932426,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004936762927790286,
|
|
"loss": 5.8342,
|
|
"mean_token_accuracy": 0.13906463459134102,
|
|
"num_tokens": 3697259.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.867275047302246,
|
|
"epoch": 1.6938981341401917,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004935825714339389,
|
|
"loss": 5.9001,
|
|
"mean_token_accuracy": 0.13478674218058587,
|
|
"num_tokens": 3708536.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.012644243240357,
|
|
"epoch": 1.6989409984871406,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004934881707291748,
|
|
"loss": 5.9065,
|
|
"mean_token_accuracy": 0.12929113656282426,
|
|
"num_tokens": 3719181.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.826500463485718,
|
|
"epoch": 1.7039838628340898,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004933930909581362,
|
|
"loss": 5.7538,
|
|
"mean_token_accuracy": 0.13596596643328668,
|
|
"num_tokens": 3729325.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.904290437698364,
|
|
"epoch": 1.7090267271810389,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004932973324163328,
|
|
"loss": 5.8745,
|
|
"mean_token_accuracy": 0.13066629618406295,
|
|
"num_tokens": 3740462.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.937609672546387,
|
|
"epoch": 1.7140695915279878,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004932008954013846,
|
|
"loss": 5.9797,
|
|
"mean_token_accuracy": 0.12895917668938636,
|
|
"num_tokens": 3752523.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.854755926132202,
|
|
"epoch": 1.7191124558749369,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004931037802130201,
|
|
"loss": 5.8405,
|
|
"mean_token_accuracy": 0.13127497360110282,
|
|
"num_tokens": 3764120.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.907617568969727,
|
|
"epoch": 1.724155320221886,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004930059871530755,
|
|
"loss": 5.972,
|
|
"mean_token_accuracy": 0.12837913185358046,
|
|
"num_tokens": 3776733.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.01545033454895,
|
|
"epoch": 1.7291981845688351,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004929075165254938,
|
|
"loss": 5.9337,
|
|
"mean_token_accuracy": 0.12843016907572746,
|
|
"num_tokens": 3787799.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.798501253128052,
|
|
"epoch": 1.7342410489157842,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004928083686363241,
|
|
"loss": 5.8562,
|
|
"mean_token_accuracy": 0.1305542565882206,
|
|
"num_tokens": 3797995.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.994955348968506,
|
|
"epoch": 1.7392839132627333,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004927085437937204,
|
|
"loss": 5.9414,
|
|
"mean_token_accuracy": 0.12499688416719437,
|
|
"num_tokens": 3809580.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.964277935028076,
|
|
"epoch": 1.7443267776096822,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004926080423079404,
|
|
"loss": 5.9669,
|
|
"mean_token_accuracy": 0.13099026530981064,
|
|
"num_tokens": 3820150.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.898656749725342,
|
|
"epoch": 1.7493696419566314,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000492506864491345,
|
|
"loss": 5.813,
|
|
"mean_token_accuracy": 0.12974431589245797,
|
|
"num_tokens": 3830391.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.816258382797241,
|
|
"epoch": 1.7544125063035805,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004924050106583976,
|
|
"loss": 5.8311,
|
|
"mean_token_accuracy": 0.13523281887173652,
|
|
"num_tokens": 3842773.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.9264304637908936,
|
|
"epoch": 1.7594553706505294,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004923024811256618,
|
|
"loss": 5.8615,
|
|
"mean_token_accuracy": 0.13258160129189492,
|
|
"num_tokens": 3853646.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.904428911209107,
|
|
"epoch": 1.7644982349974785,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000492199276211802,
|
|
"loss": 5.8332,
|
|
"mean_token_accuracy": 0.1314268544316292,
|
|
"num_tokens": 3865319.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.870353078842163,
|
|
"epoch": 1.7695410993444276,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004920953962375813,
|
|
"loss": 5.9019,
|
|
"mean_token_accuracy": 0.1266103930771351,
|
|
"num_tokens": 3877107.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.8814592361450195,
|
|
"epoch": 1.7745839636913767,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004919908415258612,
|
|
"loss": 5.9306,
|
|
"mean_token_accuracy": 0.13517394736409188,
|
|
"num_tokens": 3889334.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.958156681060791,
|
|
"epoch": 1.7796268280383258,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004918856124016002,
|
|
"loss": 5.8145,
|
|
"mean_token_accuracy": 0.13674092888832093,
|
|
"num_tokens": 3899482.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.764415264129639,
|
|
"epoch": 1.784669692385275,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004917797091918526,
|
|
"loss": 5.7577,
|
|
"mean_token_accuracy": 0.1344623699784279,
|
|
"num_tokens": 3910216.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.990887451171875,
|
|
"epoch": 1.789712556732224,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004916731322257683,
|
|
"loss": 5.9112,
|
|
"mean_token_accuracy": 0.13275586888194085,
|
|
"num_tokens": 3922246.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.829149913787842,
|
|
"epoch": 1.794755421079173,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004915658818345908,
|
|
"loss": 5.8148,
|
|
"mean_token_accuracy": 0.140170219540596,
|
|
"num_tokens": 3933688.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.8959815979003904,
|
|
"epoch": 1.799798285426122,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004914579583516568,
|
|
"loss": 5.9084,
|
|
"mean_token_accuracy": 0.12661692649126052,
|
|
"num_tokens": 3944168.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.872848033905029,
|
|
"epoch": 1.804841149773071,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000491349362112395,
|
|
"loss": 5.8429,
|
|
"mean_token_accuracy": 0.13449825197458268,
|
|
"num_tokens": 3954346.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.932722806930542,
|
|
"epoch": 1.80988401412002,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000491240093454325,
|
|
"loss": 5.946,
|
|
"mean_token_accuracy": 0.13098567500710487,
|
|
"num_tokens": 3966375.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.91828465461731,
|
|
"epoch": 1.8149268784669692,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004911301527170563,
|
|
"loss": 5.8963,
|
|
"mean_token_accuracy": 0.1361001491546631,
|
|
"num_tokens": 3977030.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.834843254089355,
|
|
"epoch": 1.8199697428139183,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004910195402422873,
|
|
"loss": 5.8289,
|
|
"mean_token_accuracy": 0.1292489267885685,
|
|
"num_tokens": 3988064.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.923317813873291,
|
|
"epoch": 1.8250126071608674,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004909082563738042,
|
|
"loss": 5.863,
|
|
"mean_token_accuracy": 0.13194844424724578,
|
|
"num_tokens": 3997994.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.94038462638855,
|
|
"epoch": 1.8300554715078166,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004907963014574797,
|
|
"loss": 5.9382,
|
|
"mean_token_accuracy": 0.12465495020151138,
|
|
"num_tokens": 4008645.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.909869003295898,
|
|
"epoch": 1.8350983358547657,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004906836758412722,
|
|
"loss": 5.8625,
|
|
"mean_token_accuracy": 0.13498851880431176,
|
|
"num_tokens": 4020387.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.740508127212524,
|
|
"epoch": 1.8401412002017146,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004905703798752248,
|
|
"loss": 5.7427,
|
|
"mean_token_accuracy": 0.1360771492123604,
|
|
"num_tokens": 4032720.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.804358339309692,
|
|
"epoch": 1.8451840645486637,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000490456413911464,
|
|
"loss": 5.8135,
|
|
"mean_token_accuracy": 0.1373528338968754,
|
|
"num_tokens": 4043438.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.825810146331787,
|
|
"epoch": 1.8502269288956126,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004903417783041987,
|
|
"loss": 5.7941,
|
|
"mean_token_accuracy": 0.14117383807897568,
|
|
"num_tokens": 4054467.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.871437740325928,
|
|
"epoch": 1.8552697932425617,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004902264734097188,
|
|
"loss": 5.8019,
|
|
"mean_token_accuracy": 0.13363157361745834,
|
|
"num_tokens": 4066418.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.954784202575683,
|
|
"epoch": 1.8603126575895108,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004901104995863948,
|
|
"loss": 5.8892,
|
|
"mean_token_accuracy": 0.1288447819650173,
|
|
"num_tokens": 4078770.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.824422836303711,
|
|
"epoch": 1.86535552193646,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004899938571946759,
|
|
"loss": 5.9346,
|
|
"mean_token_accuracy": 0.129574766010046,
|
|
"num_tokens": 4089998.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.926247024536133,
|
|
"epoch": 1.870398386283409,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004898765465970894,
|
|
"loss": 5.8355,
|
|
"mean_token_accuracy": 0.13571253269910813,
|
|
"num_tokens": 4101409.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.943756437301635,
|
|
"epoch": 1.8754412506303582,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004897585681582393,
|
|
"loss": 5.9734,
|
|
"mean_token_accuracy": 0.12765001207590104,
|
|
"num_tokens": 4113598.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.809826135635376,
|
|
"epoch": 1.8804841149773073,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004896399222448053,
|
|
"loss": 5.8918,
|
|
"mean_token_accuracy": 0.13419346362352372,
|
|
"num_tokens": 4124639.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.998760461807251,
|
|
"epoch": 1.8855269793242562,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004895206092255417,
|
|
"loss": 5.9613,
|
|
"mean_token_accuracy": 0.12527743875980377,
|
|
"num_tokens": 4136280.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.878151702880859,
|
|
"epoch": 1.8905698436712053,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004894006294712761,
|
|
"loss": 5.7956,
|
|
"mean_token_accuracy": 0.12782858684659004,
|
|
"num_tokens": 4148251.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.898477172851562,
|
|
"epoch": 1.8956127080181542,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004892799833549083,
|
|
"loss": 5.8888,
|
|
"mean_token_accuracy": 0.1343490734696388,
|
|
"num_tokens": 4159298.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.876490736007691,
|
|
"epoch": 1.9006555723651033,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004891586712514091,
|
|
"loss": 5.8005,
|
|
"mean_token_accuracy": 0.13247711658477784,
|
|
"num_tokens": 4169974.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.7899970531463625,
|
|
"epoch": 1.9056984367120524,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004890366935378196,
|
|
"loss": 5.8118,
|
|
"mean_token_accuracy": 0.13953317031264306,
|
|
"num_tokens": 4181442.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.876431465148926,
|
|
"epoch": 1.9107413010590015,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004889140505932492,
|
|
"loss": 5.8458,
|
|
"mean_token_accuracy": 0.14020064026117324,
|
|
"num_tokens": 4191535.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.856980180740356,
|
|
"epoch": 1.9157841654059506,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004887907427988749,
|
|
"loss": 5.8742,
|
|
"mean_token_accuracy": 0.13323668092489244,
|
|
"num_tokens": 4201741.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.86411485671997,
|
|
"epoch": 1.9208270297528998,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004886667705379403,
|
|
"loss": 5.7619,
|
|
"mean_token_accuracy": 0.134678003937006,
|
|
"num_tokens": 4211762.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.894043588638306,
|
|
"epoch": 1.9258698940998489,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004885421341957541,
|
|
"loss": 5.8586,
|
|
"mean_token_accuracy": 0.1314342848956585,
|
|
"num_tokens": 4222322.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.892050552368164,
|
|
"epoch": 1.9309127584467978,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004884168341596888,
|
|
"loss": 5.9206,
|
|
"mean_token_accuracy": 0.13454761281609534,
|
|
"num_tokens": 4232166.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.859141111373901,
|
|
"epoch": 1.9359556227937469,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004882908708191797,
|
|
"loss": 5.8826,
|
|
"mean_token_accuracy": 0.1334269478917122,
|
|
"num_tokens": 4243651.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.893623161315918,
|
|
"epoch": 1.9409984871406958,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00048816424456572403,
|
|
"loss": 5.7962,
|
|
"mean_token_accuracy": 0.13634647578001022,
|
|
"num_tokens": 4255137.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.759242677688599,
|
|
"epoch": 1.946041351487645,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048803695579287905,
|
|
"loss": 5.8057,
|
|
"mean_token_accuracy": 0.13784050345420837,
|
|
"num_tokens": 4264904.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.833326148986816,
|
|
"epoch": 1.951084215834594,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048790900489626104,
|
|
"loss": 5.7825,
|
|
"mean_token_accuracy": 0.13507805094122888,
|
|
"num_tokens": 4275305.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.866985082626343,
|
|
"epoch": 1.9561270801815431,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004877803922735445,
|
|
"loss": 5.8532,
|
|
"mean_token_accuracy": 0.13358560726046562,
|
|
"num_tokens": 4287038.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.980012845993042,
|
|
"epoch": 1.9611699445284922,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004876511183244603,
|
|
"loss": 5.9843,
|
|
"mean_token_accuracy": 0.12907288521528243,
|
|
"num_tokens": 4298668.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.7747985363006595,
|
|
"epoch": 1.9662128088754414,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048752118345079496,
|
|
"loss": 5.7271,
|
|
"mean_token_accuracy": 0.13709074035286903,
|
|
"num_tokens": 4309429.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.872978210449219,
|
|
"epoch": 1.9712556732223905,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048739058805638905,
|
|
"loss": 5.8107,
|
|
"mean_token_accuracy": 0.1377664566040039,
|
|
"num_tokens": 4321137.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.855399703979492,
|
|
"epoch": 1.9762985375693394,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000487259332547136,
|
|
"loss": 5.8704,
|
|
"mean_token_accuracy": 0.12618953958153725,
|
|
"num_tokens": 4331249.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.832701635360718,
|
|
"epoch": 1.9813414019162885,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000487127417330981,
|
|
"loss": 5.8406,
|
|
"mean_token_accuracy": 0.1293475955724716,
|
|
"num_tokens": 4342154.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.907890939712525,
|
|
"epoch": 1.9863842662632374,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004869948428179196,
|
|
"loss": 5.891,
|
|
"mean_token_accuracy": 0.12837604135274888,
|
|
"num_tokens": 4353906.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.861940050125122,
|
|
"epoch": 1.9914271306101865,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004868616094199964,
|
|
"loss": 5.9432,
|
|
"mean_token_accuracy": 0.12661690264940262,
|
|
"num_tokens": 4366416.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.905925703048706,
|
|
"epoch": 1.9964699949571356,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000486727717551304,
|
|
"loss": 5.8438,
|
|
"mean_token_accuracy": 0.13471274226903915,
|
|
"num_tokens": 4376743.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.746378580729167,
|
|
"epoch": 2.00100857286939,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004865931676279813,
|
|
"loss": 5.6089,
|
|
"mean_token_accuracy": 0.14421919816070133,
|
|
"num_tokens": 4385379.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.844557476043701,
|
|
"epoch": 2.006051437216339,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004864579600682125,
|
|
"loss": 5.5372,
|
|
"mean_token_accuracy": 0.14184438735246657,
|
|
"num_tokens": 4396872.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.80364499092102,
|
|
"epoch": 2.011094301563288,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00048632209529222607,
|
|
"loss": 5.4561,
|
|
"mean_token_accuracy": 0.14693821966648102,
|
|
"num_tokens": 4407075.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.741332960128784,
|
|
"epoch": 2.016137165910237,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048618557372229253,
|
|
"loss": 5.4841,
|
|
"mean_token_accuracy": 0.1523932471871376,
|
|
"num_tokens": 4418020.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.016137165910237,
|
|
"eval_entropy": 5.680995989695154,
|
|
"eval_loss": 5.989192962646484,
|
|
"eval_mean_token_accuracy": 0.13356172469736055,
|
|
"eval_num_tokens": 4418020.0,
|
|
"eval_runtime": 2.6495,
|
|
"eval_samples_per_second": 1184.365,
|
|
"eval_steps_per_second": 148.329,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.766691541671753,
|
|
"epoch": 2.0211800302571863,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004860483957827243,
|
|
"loss": 5.5215,
|
|
"mean_token_accuracy": 0.14789806455373763,
|
|
"num_tokens": 4429740.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.771888685226441,
|
|
"epoch": 2.026222894604135,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004859105618998735,
|
|
"loss": 5.4892,
|
|
"mean_token_accuracy": 0.14458957314491272,
|
|
"num_tokens": 4440183.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.785090827941895,
|
|
"epoch": 2.031265758951084,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004857720725021311,
|
|
"loss": 5.4873,
|
|
"mean_token_accuracy": 0.14502768814563752,
|
|
"num_tokens": 4451750.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.645265007019043,
|
|
"epoch": 2.036308623298033,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048563292801992534,
|
|
"loss": 5.4364,
|
|
"mean_token_accuracy": 0.1487667128443718,
|
|
"num_tokens": 4462071.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.825870656967163,
|
|
"epoch": 2.0413514876449823,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004854931288857203,
|
|
"loss": 5.6289,
|
|
"mean_token_accuracy": 0.13642685860395432,
|
|
"num_tokens": 4474241.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.759660863876343,
|
|
"epoch": 2.0463943519919314,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00048535267553401505,
|
|
"loss": 5.4615,
|
|
"mean_token_accuracy": 0.1518971547484398,
|
|
"num_tokens": 4487083.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.708598232269287,
|
|
"epoch": 2.0514372163388805,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00048521156840134193,
|
|
"loss": 5.5262,
|
|
"mean_token_accuracy": 0.1527533784508705,
|
|
"num_tokens": 4497181.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.782727146148682,
|
|
"epoch": 2.0564800806858297,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000485069807926265,
|
|
"loss": 5.4769,
|
|
"mean_token_accuracy": 0.14840264469385148,
|
|
"num_tokens": 4508250.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.71816668510437,
|
|
"epoch": 2.0615229450327788,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004849273945493792,
|
|
"loss": 5.4992,
|
|
"mean_token_accuracy": 0.1487680435180664,
|
|
"num_tokens": 4519778.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.685887575149536,
|
|
"epoch": 2.066565809379728,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004847843287133084,
|
|
"loss": 5.4897,
|
|
"mean_token_accuracy": 0.14126669764518737,
|
|
"num_tokens": 4530879.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.731546783447266,
|
|
"epoch": 2.0716086737266766,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004846406108627048,
|
|
"loss": 5.5504,
|
|
"mean_token_accuracy": 0.15017888098955154,
|
|
"num_tokens": 4542560.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.62994155883789,
|
|
"epoch": 2.0766515380736257,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048449624144424665,
|
|
"loss": 5.4279,
|
|
"mean_token_accuracy": 0.15739631205797194,
|
|
"num_tokens": 4552213.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.6950929164886475,
|
|
"epoch": 2.081694402420575,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048435122090663744,
|
|
"loss": 5.4558,
|
|
"mean_token_accuracy": 0.15148370563983918,
|
|
"num_tokens": 4562798.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.685782098770142,
|
|
"epoch": 2.086737266767524,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048420554970060424,
|
|
"loss": 5.5126,
|
|
"mean_token_accuracy": 0.14563852846622466,
|
|
"num_tokens": 4574534.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.702060794830322,
|
|
"epoch": 2.091780131114473,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00048405922827889677,
|
|
"loss": 5.4945,
|
|
"mean_token_accuracy": 0.13938834443688392,
|
|
"num_tokens": 4586239.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.6921937465667725,
|
|
"epoch": 2.096822995461422,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048391225709628515,
|
|
"loss": 5.5173,
|
|
"mean_token_accuracy": 0.14917422384023665,
|
|
"num_tokens": 4598018.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.688322830200195,
|
|
"epoch": 2.1018658598083713,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004837646366095593,
|
|
"loss": 5.497,
|
|
"mean_token_accuracy": 0.14469251781702042,
|
|
"num_tokens": 4608985.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.727775382995605,
|
|
"epoch": 2.1069087241553204,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048361636727752716,
|
|
"loss": 5.5676,
|
|
"mean_token_accuracy": 0.14773706644773482,
|
|
"num_tokens": 4621103.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.673424577713012,
|
|
"epoch": 2.1119515885022695,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004834674495610131,
|
|
"loss": 5.5367,
|
|
"mean_token_accuracy": 0.14560986086726188,
|
|
"num_tokens": 4632244.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.713310432434082,
|
|
"epoch": 2.116994452849218,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048331788392285693,
|
|
"loss": 5.4587,
|
|
"mean_token_accuracy": 0.14869346618652343,
|
|
"num_tokens": 4642387.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.717536687850952,
|
|
"epoch": 2.1220373171961673,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00048316767082791205,
|
|
"loss": 5.5909,
|
|
"mean_token_accuracy": 0.14217522516846656,
|
|
"num_tokens": 4655181.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.641996097564697,
|
|
"epoch": 2.1270801815431164,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048301681074304427,
|
|
"loss": 5.4902,
|
|
"mean_token_accuracy": 0.14716131538152694,
|
|
"num_tokens": 4665664.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.635302257537842,
|
|
"epoch": 2.1321230458900655,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00048286530413713016,
|
|
"loss": 5.458,
|
|
"mean_token_accuracy": 0.15459446460008622,
|
|
"num_tokens": 4677683.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.712548017501831,
|
|
"epoch": 2.1371659102370146,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048271315148105574,
|
|
"loss": 5.4957,
|
|
"mean_token_accuracy": 0.14238945618271828,
|
|
"num_tokens": 4688891.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.68415412902832,
|
|
"epoch": 2.1422087745839637,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000482560353247715,
|
|
"loss": 5.5419,
|
|
"mean_token_accuracy": 0.14890266060829163,
|
|
"num_tokens": 4700765.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.709896850585937,
|
|
"epoch": 2.147251638930913,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004824069099120084,
|
|
"loss": 5.5606,
|
|
"mean_token_accuracy": 0.14066436737775803,
|
|
"num_tokens": 4712004.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.670287656784057,
|
|
"epoch": 2.152294503277862,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004822528219508413,
|
|
"loss": 5.446,
|
|
"mean_token_accuracy": 0.1466579720377922,
|
|
"num_tokens": 4722427.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.709214210510254,
|
|
"epoch": 2.157337367624811,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004820980898431227,
|
|
"loss": 5.5448,
|
|
"mean_token_accuracy": 0.1513879381120205,
|
|
"num_tokens": 4733815.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.6729559898376465,
|
|
"epoch": 2.1623802319717598,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004819427140697635,
|
|
"loss": 5.5056,
|
|
"mean_token_accuracy": 0.1432117849588394,
|
|
"num_tokens": 4745070.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.68441686630249,
|
|
"epoch": 2.167423096318709,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004817866951136753,
|
|
"loss": 5.5624,
|
|
"mean_token_accuracy": 0.13918726816773414,
|
|
"num_tokens": 4756066.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.7714673519134525,
|
|
"epoch": 2.172465960665658,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00048163003345976847,
|
|
"loss": 5.5832,
|
|
"mean_token_accuracy": 0.14053326919674874,
|
|
"num_tokens": 4767153.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.683536720275879,
|
|
"epoch": 2.177508825012607,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004814727295949512,
|
|
"loss": 5.4854,
|
|
"mean_token_accuracy": 0.1484683021903038,
|
|
"num_tokens": 4777554.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.639589738845825,
|
|
"epoch": 2.1825516893595562,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00048131478400812735,
|
|
"loss": 5.5327,
|
|
"mean_token_accuracy": 0.1460942767560482,
|
|
"num_tokens": 4788881.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.701530885696411,
|
|
"epoch": 2.1875945537065054,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004811561971901956,
|
|
"loss": 5.4979,
|
|
"mean_token_accuracy": 0.14612958282232286,
|
|
"num_tokens": 4798680.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.667514801025391,
|
|
"epoch": 2.1926374180534545,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004809969696340474,
|
|
"loss": 5.5044,
|
|
"mean_token_accuracy": 0.15070030689239503,
|
|
"num_tokens": 4808096.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.670957708358765,
|
|
"epoch": 2.1976802824004036,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048083710183456554,
|
|
"loss": 5.4629,
|
|
"mean_token_accuracy": 0.15089642405509948,
|
|
"num_tokens": 4818778.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.57321720123291,
|
|
"epoch": 2.2027231467473527,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048067659428862306,
|
|
"loss": 5.4543,
|
|
"mean_token_accuracy": 0.15111969113349916,
|
|
"num_tokens": 4828586.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.672693300247192,
|
|
"epoch": 2.2077660110943014,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048051544749508096,
|
|
"loss": 5.5658,
|
|
"mean_token_accuracy": 0.14453586041927338,
|
|
"num_tokens": 4839419.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.666914653778076,
|
|
"epoch": 2.2128088754412505,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004803536619547872,
|
|
"loss": 5.4861,
|
|
"mean_token_accuracy": 0.15151002258062363,
|
|
"num_tokens": 4849700.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.6714146614074705,
|
|
"epoch": 2.2178517397881996,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000480191238170575,
|
|
"loss": 5.6584,
|
|
"mean_token_accuracy": 0.13440413549542427,
|
|
"num_tokens": 4860261.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.758241844177246,
|
|
"epoch": 2.2228946041351487,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004800281766472612,
|
|
"loss": 5.5626,
|
|
"mean_token_accuracy": 0.14456569850444795,
|
|
"num_tokens": 4872277.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.672854328155518,
|
|
"epoch": 2.227937468482098,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00047986447789164486,
|
|
"loss": 5.6228,
|
|
"mean_token_accuracy": 0.1488087370991707,
|
|
"num_tokens": 4883512.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.6741656303405765,
|
|
"epoch": 2.232980332829047,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004797001424125054,
|
|
"loss": 5.5327,
|
|
"mean_token_accuracy": 0.14639116078615189,
|
|
"num_tokens": 4896197.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.732757711410523,
|
|
"epoch": 2.238023197175996,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00047953517072060124,
|
|
"loss": 5.4931,
|
|
"mean_token_accuracy": 0.15128834694623947,
|
|
"num_tokens": 4907397.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.637088871002197,
|
|
"epoch": 2.243066061522945,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004793695633286684,
|
|
"loss": 5.5785,
|
|
"mean_token_accuracy": 0.1417090468108654,
|
|
"num_tokens": 4919008.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.6243401050567625,
|
|
"epoch": 2.2481089258698943,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047920332075141825,
|
|
"loss": 5.4902,
|
|
"mean_token_accuracy": 0.14776550382375717,
|
|
"num_tokens": 4930877.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.6722338676452635,
|
|
"epoch": 2.253151790216843,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004790364435055367,
|
|
"loss": 5.3959,
|
|
"mean_token_accuracy": 0.15714615434408188,
|
|
"num_tokens": 4942274.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.6240647315979,
|
|
"epoch": 2.258194654563792,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000478868932109682,
|
|
"loss": 5.491,
|
|
"mean_token_accuracy": 0.14969139397144318,
|
|
"num_tokens": 4953740.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.6926532745361325,
|
|
"epoch": 2.263237518910741,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004787007870844835,
|
|
"loss": 5.5324,
|
|
"mean_token_accuracy": 0.14889150112867355,
|
|
"num_tokens": 4964760.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.6714859962463375,
|
|
"epoch": 2.2682803832576903,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00047853200895253976,
|
|
"loss": 5.5913,
|
|
"mean_token_accuracy": 0.1424285888671875,
|
|
"num_tokens": 4976404.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.653643035888672,
|
|
"epoch": 2.2733232476046394,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00047836259823841716,
|
|
"loss": 5.4754,
|
|
"mean_token_accuracy": 0.14482759982347487,
|
|
"num_tokens": 4987552.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.630600309371948,
|
|
"epoch": 2.2783661119515886,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047819255546864797,
|
|
"loss": 5.487,
|
|
"mean_token_accuracy": 0.1477342277765274,
|
|
"num_tokens": 4998912.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.682914304733276,
|
|
"epoch": 2.2834089762985377,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004780218811717293,
|
|
"loss": 5.4651,
|
|
"mean_token_accuracy": 0.15380304604768752,
|
|
"num_tokens": 5009556.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.624031257629395,
|
|
"epoch": 2.288451840645487,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047785057587812057,
|
|
"loss": 5.5095,
|
|
"mean_token_accuracy": 0.14924681186676025,
|
|
"num_tokens": 5021569.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.7075849056243895,
|
|
"epoch": 2.293494704992436,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047767864012024266,
|
|
"loss": 5.58,
|
|
"mean_token_accuracy": 0.14542651996016503,
|
|
"num_tokens": 5032994.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.7110589981079105,
|
|
"epoch": 2.2985375693393846,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047750607443247585,
|
|
"loss": 5.5193,
|
|
"mean_token_accuracy": 0.14738448411226274,
|
|
"num_tokens": 5044681.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.618875551223755,
|
|
"epoch": 2.3035804336863337,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00047733287935115837,
|
|
"loss": 5.4435,
|
|
"mean_token_accuracy": 0.15486137419939042,
|
|
"num_tokens": 5055509.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.519594621658325,
|
|
"epoch": 2.308623298033283,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004771590554145844,
|
|
"loss": 5.4787,
|
|
"mean_token_accuracy": 0.15279970616102218,
|
|
"num_tokens": 5066903.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.660610771179199,
|
|
"epoch": 2.313666162380232,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047698460316300285,
|
|
"loss": 5.5231,
|
|
"mean_token_accuracy": 0.14691229909658432,
|
|
"num_tokens": 5078411.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.646321582794189,
|
|
"epoch": 2.318709026727181,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004768095231386153,
|
|
"loss": 5.5308,
|
|
"mean_token_accuracy": 0.14467755407094957,
|
|
"num_tokens": 5089811.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.688470268249512,
|
|
"epoch": 2.32375189107413,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004766338158855744,
|
|
"loss": 5.5323,
|
|
"mean_token_accuracy": 0.15395605564117432,
|
|
"num_tokens": 5101196.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.576445579528809,
|
|
"epoch": 2.3287947554210793,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004764574819499826,
|
|
"loss": 5.5122,
|
|
"mean_token_accuracy": 0.150606869161129,
|
|
"num_tokens": 5112379.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.704416084289551,
|
|
"epoch": 2.3338376197680284,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047628052187988955,
|
|
"loss": 5.6262,
|
|
"mean_token_accuracy": 0.13457250222563744,
|
|
"num_tokens": 5125065.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.766238880157471,
|
|
"epoch": 2.338880484114977,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004761029362252914,
|
|
"loss": 5.5993,
|
|
"mean_token_accuracy": 0.14355267137289046,
|
|
"num_tokens": 5137471.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.736388301849365,
|
|
"epoch": 2.343923348461926,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047592472553812856,
|
|
"loss": 5.5429,
|
|
"mean_token_accuracy": 0.15075437873601913,
|
|
"num_tokens": 5147682.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.644363355636597,
|
|
"epoch": 2.3489662128088753,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004757458903722837,
|
|
"loss": 5.5152,
|
|
"mean_token_accuracy": 0.14494604468345643,
|
|
"num_tokens": 5156143.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.69903416633606,
|
|
"epoch": 2.3540090771558244,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00047556643128358075,
|
|
"loss": 5.5571,
|
|
"mean_token_accuracy": 0.14494485259056092,
|
|
"num_tokens": 5167462.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.586874771118164,
|
|
"epoch": 2.3590519415027735,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047538634882978277,
|
|
"loss": 5.4811,
|
|
"mean_token_accuracy": 0.1485794246196747,
|
|
"num_tokens": 5178030.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.604873609542847,
|
|
"epoch": 2.3640948058497226,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047520564357059013,
|
|
"loss": 5.5169,
|
|
"mean_token_accuracy": 0.1468599297106266,
|
|
"num_tokens": 5188296.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.548481798171997,
|
|
"epoch": 2.3691376701966718,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004750243160676389,
|
|
"loss": 5.4911,
|
|
"mean_token_accuracy": 0.1502973645925522,
|
|
"num_tokens": 5199058.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.66943097114563,
|
|
"epoch": 2.374180534543621,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004748423668844992,
|
|
"loss": 5.4603,
|
|
"mean_token_accuracy": 0.14912888109683992,
|
|
"num_tokens": 5209314.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.676968717575074,
|
|
"epoch": 2.37922339889057,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004746597965866733,
|
|
"loss": 5.588,
|
|
"mean_token_accuracy": 0.14606355130672455,
|
|
"num_tokens": 5220248.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.6687804698944095,
|
|
"epoch": 2.384266263237519,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004744766057415938,
|
|
"loss": 5.546,
|
|
"mean_token_accuracy": 0.13689773082733153,
|
|
"num_tokens": 5230874.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.654182243347168,
|
|
"epoch": 2.389309127584468,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004742927949186222,
|
|
"loss": 5.5029,
|
|
"mean_token_accuracy": 0.15400156527757644,
|
|
"num_tokens": 5243328.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.560733842849731,
|
|
"epoch": 2.394351991931417,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00047410836468904664,
|
|
"loss": 5.4458,
|
|
"mean_token_accuracy": 0.14656009376049042,
|
|
"num_tokens": 5254357.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.561968517303467,
|
|
"epoch": 2.399394856278366,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047392331562608073,
|
|
"loss": 5.4535,
|
|
"mean_token_accuracy": 0.15026493668556212,
|
|
"num_tokens": 5264492.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.575039291381836,
|
|
"epoch": 2.404437720625315,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004737376483048609,
|
|
"loss": 5.4142,
|
|
"mean_token_accuracy": 0.15389057099819184,
|
|
"num_tokens": 5274816.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.613295602798462,
|
|
"epoch": 2.4094805849722642,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004735513633024458,
|
|
"loss": 5.5165,
|
|
"mean_token_accuracy": 0.1459888331592083,
|
|
"num_tokens": 5286855.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.683413743972778,
|
|
"epoch": 2.4145234493192134,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004733644611978134,
|
|
"loss": 5.6262,
|
|
"mean_token_accuracy": 0.13750532865524293,
|
|
"num_tokens": 5299594.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.579679918289185,
|
|
"epoch": 2.4195663136661625,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004731769425718597,
|
|
"loss": 5.4423,
|
|
"mean_token_accuracy": 0.1451920449733734,
|
|
"num_tokens": 5310078.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.584488201141357,
|
|
"epoch": 2.4246091780131116,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000472988808007397,
|
|
"loss": 5.4762,
|
|
"mean_token_accuracy": 0.15450869500637054,
|
|
"num_tokens": 5319544.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.617080354690552,
|
|
"epoch": 2.4296520423600603,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00047280005808915176,
|
|
"loss": 5.5489,
|
|
"mean_token_accuracy": 0.14544432386755943,
|
|
"num_tokens": 5331025.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.547574234008789,
|
|
"epoch": 2.4346949067070094,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004726106934037631,
|
|
"loss": 5.5234,
|
|
"mean_token_accuracy": 0.15173332393169403,
|
|
"num_tokens": 5343086.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.712149143218994,
|
|
"epoch": 2.4397377710539585,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004724207145397809,
|
|
"loss": 5.5559,
|
|
"mean_token_accuracy": 0.15391047894954682,
|
|
"num_tokens": 5353598.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.608832216262817,
|
|
"epoch": 2.4447806354009076,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047223012208766384,
|
|
"loss": 5.4557,
|
|
"mean_token_accuracy": 0.14823706597089767,
|
|
"num_tokens": 5363607.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.574588060379028,
|
|
"epoch": 2.4498234997478567,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004720389166397777,
|
|
"loss": 5.5194,
|
|
"mean_token_accuracy": 0.14685906171798707,
|
|
"num_tokens": 5375198.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.589098119735718,
|
|
"epoch": 2.454866364094806,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00047184709879039325,
|
|
"loss": 5.4897,
|
|
"mean_token_accuracy": 0.1493762269616127,
|
|
"num_tokens": 5386058.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.667830181121826,
|
|
"epoch": 2.459909228441755,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047165466913568497,
|
|
"loss": 5.5167,
|
|
"mean_token_accuracy": 0.14955587536096573,
|
|
"num_tokens": 5395720.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.5903956413269045,
|
|
"epoch": 2.464952092788704,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004714616282737285,
|
|
"loss": 5.5626,
|
|
"mean_token_accuracy": 0.1448132574558258,
|
|
"num_tokens": 5406252.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.6490295886993405,
|
|
"epoch": 2.469994957135653,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004712679768044995,
|
|
"loss": 5.5611,
|
|
"mean_token_accuracy": 0.14693916216492653,
|
|
"num_tokens": 5418125.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.699648809432984,
|
|
"epoch": 2.4750378214826023,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000471073715329871,
|
|
"loss": 5.5317,
|
|
"mean_token_accuracy": 0.15424191057682038,
|
|
"num_tokens": 5428095.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.671195602416992,
|
|
"epoch": 2.4800806858295514,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004708788444536123,
|
|
"loss": 5.5417,
|
|
"mean_token_accuracy": 0.14158782064914704,
|
|
"num_tokens": 5438359.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.6120024681091305,
|
|
"epoch": 2.4851235501765,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004706833647813865,
|
|
"loss": 5.4734,
|
|
"mean_token_accuracy": 0.15623558014631272,
|
|
"num_tokens": 5448559.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.546583366394043,
|
|
"epoch": 2.4901664145234492,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00047048727692074893,
|
|
"loss": 5.4917,
|
|
"mean_token_accuracy": 0.14745179563760757,
|
|
"num_tokens": 5459739.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.572244501113891,
|
|
"epoch": 2.4952092788703983,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004702905814811452,
|
|
"loss": 5.4666,
|
|
"mean_token_accuracy": 0.1457386389374733,
|
|
"num_tokens": 5471160.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.528280639648438,
|
|
"epoch": 2.5002521432173475,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00047009327907390926,
|
|
"loss": 5.5297,
|
|
"mean_token_accuracy": 0.14906351193785666,
|
|
"num_tokens": 5482344.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.677863693237304,
|
|
"epoch": 2.5052950075642966,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004698953703122616,
|
|
"loss": 5.5462,
|
|
"mean_token_accuracy": 0.14721935838460923,
|
|
"num_tokens": 5493111.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.483181524276733,
|
|
"epoch": 2.5103378719112457,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046969685581130715,
|
|
"loss": 5.466,
|
|
"mean_token_accuracy": 0.1502784177660942,
|
|
"num_tokens": 5503877.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.6861711025238035,
|
|
"epoch": 2.515380736258195,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00046949773618803353,
|
|
"loss": 5.4765,
|
|
"mean_token_accuracy": 0.14919823929667472,
|
|
"num_tokens": 5514067.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.66523585319519,
|
|
"epoch": 2.5204236006051435,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00046929801206130925,
|
|
"loss": 5.5121,
|
|
"mean_token_accuracy": 0.14804998338222503,
|
|
"num_tokens": 5525534.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.5204236006051435,
|
|
"eval_entropy": 5.525525642715338,
|
|
"eval_loss": 5.9093546867370605,
|
|
"eval_mean_token_accuracy": 0.1372140715936668,
|
|
"eval_num_tokens": 5525534.0,
|
|
"eval_runtime": 2.8743,
|
|
"eval_samples_per_second": 1091.732,
|
|
"eval_steps_per_second": 136.727,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.593466424942017,
|
|
"epoch": 2.5254664649520926,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046909768405188135,
|
|
"loss": 5.495,
|
|
"mean_token_accuracy": 0.15152756124734879,
|
|
"num_tokens": 5535525.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.551295900344849,
|
|
"epoch": 2.5305093292990417,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046889675278237404,
|
|
"loss": 5.5033,
|
|
"mean_token_accuracy": 0.15482114404439926,
|
|
"num_tokens": 5546129.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.730327129364014,
|
|
"epoch": 2.535552193645991,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004686952188772863,
|
|
"loss": 5.6314,
|
|
"mean_token_accuracy": 0.14667745381593705,
|
|
"num_tokens": 5557548.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.676825952529907,
|
|
"epoch": 2.54059505799294,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00046849308296299005,
|
|
"loss": 5.5418,
|
|
"mean_token_accuracy": 0.15349968373775483,
|
|
"num_tokens": 5567367.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.651474905014038,
|
|
"epoch": 2.545637922339889,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004682903456677285,
|
|
"loss": 5.4609,
|
|
"mean_token_accuracy": 0.1505977191030979,
|
|
"num_tokens": 5577489.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.576373434066772,
|
|
"epoch": 2.550680786686838,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046808700762161385,
|
|
"loss": 5.5061,
|
|
"mean_token_accuracy": 0.1442405879497528,
|
|
"num_tokens": 5587877.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.603444576263428,
|
|
"epoch": 2.5557236510337873,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004678830694566253,
|
|
"loss": 5.5106,
|
|
"mean_token_accuracy": 0.15525266379117966,
|
|
"num_tokens": 5597863.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.572502756118775,
|
|
"epoch": 2.5607665153807364,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004676785318066075,
|
|
"loss": 5.5096,
|
|
"mean_token_accuracy": 0.14986335039138793,
|
|
"num_tokens": 5608141.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.598837566375733,
|
|
"epoch": 2.5658093797276855,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004674733953072682,
|
|
"loss": 5.5944,
|
|
"mean_token_accuracy": 0.1454719938337803,
|
|
"num_tokens": 5619660.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.733735609054565,
|
|
"epoch": 2.5708522440746346,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004672676605961763,
|
|
"loss": 5.5688,
|
|
"mean_token_accuracy": 0.1414414018392563,
|
|
"num_tokens": 5630271.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.620486211776734,
|
|
"epoch": 2.5758951084215833,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00046706132831276007,
|
|
"loss": 5.5717,
|
|
"mean_token_accuracy": 0.14866012409329415,
|
|
"num_tokens": 5642594.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.5966949462890625,
|
|
"epoch": 2.5809379727685324,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046685439909830486,
|
|
"loss": 5.4977,
|
|
"mean_token_accuracy": 0.15163239985704421,
|
|
"num_tokens": 5653361.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.535309505462647,
|
|
"epoch": 2.5859808371154815,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004666468735959516,
|
|
"loss": 5.4287,
|
|
"mean_token_accuracy": 0.1503990426659584,
|
|
"num_tokens": 5664167.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.582321882247925,
|
|
"epoch": 2.5910237014624307,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004664387524506943,
|
|
"loss": 5.5081,
|
|
"mean_token_accuracy": 0.14768971651792526,
|
|
"num_tokens": 5675903.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.615547752380371,
|
|
"epoch": 2.59606656580938,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004662300363093784,
|
|
"loss": 5.4954,
|
|
"mean_token_accuracy": 0.14742980152368546,
|
|
"num_tokens": 5686319.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.685475921630859,
|
|
"epoch": 2.601109430156329,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004660207258206983,
|
|
"loss": 5.6048,
|
|
"mean_token_accuracy": 0.14207042753696442,
|
|
"num_tokens": 5696000.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.599239778518677,
|
|
"epoch": 2.606152294503278,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00046581082163519585,
|
|
"loss": 5.5057,
|
|
"mean_token_accuracy": 0.14740608334541322,
|
|
"num_tokens": 5708145.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.531191062927246,
|
|
"epoch": 2.6111951588502267,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004656003244052582,
|
|
"loss": 5.4622,
|
|
"mean_token_accuracy": 0.15142022371292113,
|
|
"num_tokens": 5718053.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.606213474273682,
|
|
"epoch": 2.616238023197176,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004653892347851155,
|
|
"loss": 5.5347,
|
|
"mean_token_accuracy": 0.14673714712262154,
|
|
"num_tokens": 5728723.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.530488300323486,
|
|
"epoch": 2.621280887544125,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004651775534308392,
|
|
"loss": 5.4554,
|
|
"mean_token_accuracy": 0.1564555361866951,
|
|
"num_tokens": 5739414.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.658373689651489,
|
|
"epoch": 2.626323751891074,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004649652810003398,
|
|
"loss": 5.6026,
|
|
"mean_token_accuracy": 0.14294807687401773,
|
|
"num_tokens": 5751247.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.584903287887573,
|
|
"epoch": 2.631366616238023,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00046475241815336504,
|
|
"loss": 5.4593,
|
|
"mean_token_accuracy": 0.1572932004928589,
|
|
"num_tokens": 5761791.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.579724216461182,
|
|
"epoch": 2.6364094805849723,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004645389655514974,
|
|
"loss": 5.4536,
|
|
"mean_token_accuracy": 0.1468845620751381,
|
|
"num_tokens": 5773597.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.573400163650513,
|
|
"epoch": 2.6414523449319214,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004643249238581525,
|
|
"loss": 5.4724,
|
|
"mean_token_accuracy": 0.14879051074385644,
|
|
"num_tokens": 5782963.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.548124313354492,
|
|
"epoch": 2.6464952092788705,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004641102937385769,
|
|
"loss": 5.5172,
|
|
"mean_token_accuracy": 0.144728322327137,
|
|
"num_tokens": 5793474.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.596257877349854,
|
|
"epoch": 2.6515380736258196,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004638950758598461,
|
|
"loss": 5.4988,
|
|
"mean_token_accuracy": 0.1495629146695137,
|
|
"num_tokens": 5804018.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.678882837295532,
|
|
"epoch": 2.6565809379727687,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046367927089086203,
|
|
"loss": 5.526,
|
|
"mean_token_accuracy": 0.1477683201432228,
|
|
"num_tokens": 5815897.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.595109319686889,
|
|
"epoch": 2.661623802319718,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046346287950235163,
|
|
"loss": 5.5415,
|
|
"mean_token_accuracy": 0.1455786257982254,
|
|
"num_tokens": 5825883.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.632021903991699,
|
|
"epoch": 2.6666666666666665,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004632459023668642,
|
|
"loss": 5.5307,
|
|
"mean_token_accuracy": 0.14996212124824523,
|
|
"num_tokens": 5836668.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.66527214050293,
|
|
"epoch": 2.6717095310136156,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004630283401587699,
|
|
"loss": 5.6498,
|
|
"mean_token_accuracy": 0.13901092410087584,
|
|
"num_tokens": 5848497.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.588075113296509,
|
|
"epoch": 2.6767523953605648,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004628101935542568,
|
|
"loss": 5.4501,
|
|
"mean_token_accuracy": 0.15287178307771682,
|
|
"num_tokens": 5859231.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.612195348739624,
|
|
"epoch": 2.681795259707514,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046259146323132986,
|
|
"loss": 5.6019,
|
|
"mean_token_accuracy": 0.14820815920829772,
|
|
"num_tokens": 5871190.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.575901842117309,
|
|
"epoch": 2.686838124054463,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046237214986980776,
|
|
"loss": 5.4655,
|
|
"mean_token_accuracy": 0.1501447468996048,
|
|
"num_tokens": 5881788.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.587506818771362,
|
|
"epoch": 2.691880988401412,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004621522541513216,
|
|
"loss": 5.4506,
|
|
"mean_token_accuracy": 0.15079249441623688,
|
|
"num_tokens": 5891289.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.564803504943848,
|
|
"epoch": 2.696923852748361,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004619317767593123,
|
|
"loss": 5.4942,
|
|
"mean_token_accuracy": 0.14542635530233383,
|
|
"num_tokens": 5903594.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.707402515411377,
|
|
"epoch": 2.70196671709531,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00046171071837902873,
|
|
"loss": 5.6047,
|
|
"mean_token_accuracy": 0.1481708064675331,
|
|
"num_tokens": 5914592.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.53761305809021,
|
|
"epoch": 2.707009581442259,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004614890796975254,
|
|
"loss": 5.4338,
|
|
"mean_token_accuracy": 0.15292296111583709,
|
|
"num_tokens": 5927137.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.565995931625366,
|
|
"epoch": 2.712052445789208,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004612668614036605,
|
|
"loss": 5.549,
|
|
"mean_token_accuracy": 0.14784384816884993,
|
|
"num_tokens": 5938730.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.633968448638916,
|
|
"epoch": 2.7170953101361572,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004610440641880934,
|
|
"loss": 5.5492,
|
|
"mean_token_accuracy": 0.14645204693078995,
|
|
"num_tokens": 5948912.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.572774600982666,
|
|
"epoch": 2.7221381744831064,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046082068874328324,
|
|
"loss": 5.4858,
|
|
"mean_token_accuracy": 0.15176114588975906,
|
|
"num_tokens": 5960253.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.649148559570312,
|
|
"epoch": 2.7271810388300555,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046059673576348596,
|
|
"loss": 5.5845,
|
|
"mean_token_accuracy": 0.14117211401462554,
|
|
"num_tokens": 5972615.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.6746245384216305,
|
|
"epoch": 2.7322239031770046,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00046037220594475264,
|
|
"loss": 5.5468,
|
|
"mean_token_accuracy": 0.15068346709012986,
|
|
"num_tokens": 5985023.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.565754175186157,
|
|
"epoch": 2.7372667675239537,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000460147099984927,
|
|
"loss": 5.4541,
|
|
"mean_token_accuracy": 0.15249723494052886,
|
|
"num_tokens": 5996445.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.593200969696045,
|
|
"epoch": 2.742309631870903,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004599214185836438,
|
|
"loss": 5.5025,
|
|
"mean_token_accuracy": 0.14778674840927125,
|
|
"num_tokens": 6007385.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.548349905014038,
|
|
"epoch": 2.747352496217852,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00045969516244232574,
|
|
"loss": 5.5312,
|
|
"mean_token_accuracy": 0.15147375911474228,
|
|
"num_tokens": 6018174.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.5658186912536625,
|
|
"epoch": 2.752395360564801,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045946833226418235,
|
|
"loss": 5.4695,
|
|
"mean_token_accuracy": 0.14571216255426406,
|
|
"num_tokens": 6030049.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.591985177993775,
|
|
"epoch": 2.7574382249117497,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004592409287542069,
|
|
"loss": 5.4939,
|
|
"mean_token_accuracy": 0.14509734213352204,
|
|
"num_tokens": 6040155.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.527977037429809,
|
|
"epoch": 2.762481089258699,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004590129526191748,
|
|
"loss": 5.5031,
|
|
"mean_token_accuracy": 0.1559046670794487,
|
|
"num_tokens": 6050569.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.617687463760376,
|
|
"epoch": 2.767523953605648,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004587844045676413,
|
|
"loss": 5.4514,
|
|
"mean_token_accuracy": 0.15295047760009767,
|
|
"num_tokens": 6060871.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.529979562759399,
|
|
"epoch": 2.772566817952597,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045855528530993874,
|
|
"loss": 5.4697,
|
|
"mean_token_accuracy": 0.15294987559318543,
|
|
"num_tokens": 6073045.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.5869945049285885,
|
|
"epoch": 2.777609682299546,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045832559555817525,
|
|
"loss": 5.5101,
|
|
"mean_token_accuracy": 0.1468508929014206,
|
|
"num_tokens": 6084798.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.612193822860718,
|
|
"epoch": 2.7826525466464953,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045809533602623185,
|
|
"loss": 5.4468,
|
|
"mean_token_accuracy": 0.14791919142007828,
|
|
"num_tokens": 6095831.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.583421993255615,
|
|
"epoch": 2.7876954109934444,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004578645074297603,
|
|
"loss": 5.4261,
|
|
"mean_token_accuracy": 0.15392916053533554,
|
|
"num_tokens": 6106958.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.504717397689819,
|
|
"epoch": 2.792738275340393,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045763311048618144,
|
|
"loss": 5.4444,
|
|
"mean_token_accuracy": 0.15830655097961427,
|
|
"num_tokens": 6118165.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.51776533126831,
|
|
"epoch": 2.797781139687342,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004574011459146822,
|
|
"loss": 5.454,
|
|
"mean_token_accuracy": 0.1573981299996376,
|
|
"num_tokens": 6129297.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.621445226669311,
|
|
"epoch": 2.8028240040342913,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00045716861443621386,
|
|
"loss": 5.4746,
|
|
"mean_token_accuracy": 0.1501913473010063,
|
|
"num_tokens": 6140071.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.585222578048706,
|
|
"epoch": 2.8078668683812404,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045693551677348967,
|
|
"loss": 5.4878,
|
|
"mean_token_accuracy": 0.15021382346749307,
|
|
"num_tokens": 6150940.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.570693302154541,
|
|
"epoch": 2.8129097327281896,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004567018536509823,
|
|
"loss": 5.4688,
|
|
"mean_token_accuracy": 0.14465128928422927,
|
|
"num_tokens": 6161966.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.64210352897644,
|
|
"epoch": 2.8179525970751387,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045646762579492245,
|
|
"loss": 5.5381,
|
|
"mean_token_accuracy": 0.14720866531133653,
|
|
"num_tokens": 6174455.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.533165168762207,
|
|
"epoch": 2.822995461422088,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045623283393329546,
|
|
"loss": 5.458,
|
|
"mean_token_accuracy": 0.14846094697713852,
|
|
"num_tokens": 6184789.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.582577800750732,
|
|
"epoch": 2.828038325769037,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045599747879584,
|
|
"loss": 5.5512,
|
|
"mean_token_accuracy": 0.14493268132209777,
|
|
"num_tokens": 6195704.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.622770500183106,
|
|
"epoch": 2.833081190115986,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004557615611140451,
|
|
"loss": 5.434,
|
|
"mean_token_accuracy": 0.15145912021398544,
|
|
"num_tokens": 6206290.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.586861419677734,
|
|
"epoch": 2.838124054462935,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004555250816211485,
|
|
"loss": 5.5501,
|
|
"mean_token_accuracy": 0.14370328187942505,
|
|
"num_tokens": 6216660.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.588488483428955,
|
|
"epoch": 2.8431669188098843,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00045528804105213395,
|
|
"loss": 5.4789,
|
|
"mean_token_accuracy": 0.1566378131508827,
|
|
"num_tokens": 6228005.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.521516418457031,
|
|
"epoch": 2.848209783156833,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004550504401437289,
|
|
"loss": 5.4483,
|
|
"mean_token_accuracy": 0.14785733968019485,
|
|
"num_tokens": 6239447.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.578975820541382,
|
|
"epoch": 2.853252647503782,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00045481227963440235,
|
|
"loss": 5.5596,
|
|
"mean_token_accuracy": 0.14091725498437882,
|
|
"num_tokens": 6252312.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.599402666091919,
|
|
"epoch": 2.858295511850731,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045457356026436295,
|
|
"loss": 5.4868,
|
|
"mean_token_accuracy": 0.14786000698804855,
|
|
"num_tokens": 6264131.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.475321531295776,
|
|
"epoch": 2.8633383761976803,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045433428277555577,
|
|
"loss": 5.4332,
|
|
"mean_token_accuracy": 0.15369775146245956,
|
|
"num_tokens": 6275161.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.5388298511505125,
|
|
"epoch": 2.8683812405446294,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004540944479116609,
|
|
"loss": 5.4705,
|
|
"mean_token_accuracy": 0.14685434401035308,
|
|
"num_tokens": 6285655.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.6022858142852785,
|
|
"epoch": 2.8734241048915785,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004538540564180906,
|
|
"loss": 5.5786,
|
|
"mean_token_accuracy": 0.15059732943773269,
|
|
"num_tokens": 6296532.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.699536323547363,
|
|
"epoch": 2.8784669692385276,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004536131090419871,
|
|
"loss": 5.6197,
|
|
"mean_token_accuracy": 0.1454712927341461,
|
|
"num_tokens": 6308229.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.587550020217895,
|
|
"epoch": 2.8835098335854763,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004533716065322205,
|
|
"loss": 5.5321,
|
|
"mean_token_accuracy": 0.15190866738557815,
|
|
"num_tokens": 6321257.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.63558349609375,
|
|
"epoch": 2.8885526979324254,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004531295496393861,
|
|
"loss": 5.4879,
|
|
"mean_token_accuracy": 0.14733844846487046,
|
|
"num_tokens": 6332040.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.500462245941162,
|
|
"epoch": 2.8935955622793745,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004528869391158023,
|
|
"loss": 5.3662,
|
|
"mean_token_accuracy": 0.15406107902526855,
|
|
"num_tokens": 6342925.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.568462133407593,
|
|
"epoch": 2.8986384266263236,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00045264377571550834,
|
|
"loss": 5.5423,
|
|
"mean_token_accuracy": 0.14780747443437575,
|
|
"num_tokens": 6354263.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.69741268157959,
|
|
"epoch": 2.9036812909732728,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004524000601942615,
|
|
"loss": 5.5499,
|
|
"mean_token_accuracy": 0.1434149533510208,
|
|
"num_tokens": 6365497.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.624714279174805,
|
|
"epoch": 2.908724155320222,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004521557933095354,
|
|
"loss": 5.589,
|
|
"mean_token_accuracy": 0.14008525684475898,
|
|
"num_tokens": 6376501.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.6361486434936525,
|
|
"epoch": 2.913767019667171,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000451910975820517,
|
|
"loss": 5.5345,
|
|
"mean_token_accuracy": 0.14242256581783294,
|
|
"num_tokens": 6387632.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.517276191711426,
|
|
"epoch": 2.91880988401412,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045166560848810476,
|
|
"loss": 5.5235,
|
|
"mean_token_accuracy": 0.1575486272573471,
|
|
"num_tokens": 6398813.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.664028882980347,
|
|
"epoch": 2.9238527483610692,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045141969207490606,
|
|
"loss": 5.4292,
|
|
"mean_token_accuracy": 0.15005266070365905,
|
|
"num_tokens": 6409532.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.5593585014343265,
|
|
"epoch": 2.9288956127080183,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004511732273452347,
|
|
"loss": 5.5639,
|
|
"mean_token_accuracy": 0.14502508342266082,
|
|
"num_tokens": 6420755.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.6420892715454105,
|
|
"epoch": 2.9339384770549675,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004509262150651089,
|
|
"loss": 5.5781,
|
|
"mean_token_accuracy": 0.15124068707227706,
|
|
"num_tokens": 6430662.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.609274435043335,
|
|
"epoch": 2.938981341401916,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045067865600224833,
|
|
"loss": 5.4523,
|
|
"mean_token_accuracy": 0.15559796392917633,
|
|
"num_tokens": 6441230.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.6485778331756595,
|
|
"epoch": 2.9440242057488653,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004504305509260724,
|
|
"loss": 5.6092,
|
|
"mean_token_accuracy": 0.1395176500082016,
|
|
"num_tokens": 6452574.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.579356575012207,
|
|
"epoch": 2.9490670700958144,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004501819006076974,
|
|
"loss": 5.4658,
|
|
"mean_token_accuracy": 0.1548192694783211,
|
|
"num_tokens": 6462562.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.535385990142823,
|
|
"epoch": 2.9541099344427635,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00044993270581993413,
|
|
"loss": 5.4638,
|
|
"mean_token_accuracy": 0.15147760957479478,
|
|
"num_tokens": 6473594.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.600867652893067,
|
|
"epoch": 2.9591527987897126,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00044968296733728587,
|
|
"loss": 5.5028,
|
|
"mean_token_accuracy": 0.15180351287126542,
|
|
"num_tokens": 6484129.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.627963018417359,
|
|
"epoch": 2.9641956631366617,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004494326859359456,
|
|
"loss": 5.474,
|
|
"mean_token_accuracy": 0.14665171951055528,
|
|
"num_tokens": 6495584.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.542156982421875,
|
|
"epoch": 2.969238527483611,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044918186239379363,
|
|
"loss": 5.5546,
|
|
"mean_token_accuracy": 0.15177774727344512,
|
|
"num_tokens": 6507872.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.545723962783813,
|
|
"epoch": 2.9742813918305595,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004489304974903953,
|
|
"loss": 5.4256,
|
|
"mean_token_accuracy": 0.15348378866910933,
|
|
"num_tokens": 6519172.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.558694553375244,
|
|
"epoch": 2.9793242561775086,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004486785920069985,
|
|
"loss": 5.5363,
|
|
"mean_token_accuracy": 0.15182727128267287,
|
|
"num_tokens": 6530151.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.6444190502166744,
|
|
"epoch": 2.9843671205244577,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004484261467265313,
|
|
"loss": 5.5246,
|
|
"mean_token_accuracy": 0.14669964984059333,
|
|
"num_tokens": 6541881.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.59735074043274,
|
|
"epoch": 2.989409984871407,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00044817316243359956,
|
|
"loss": 5.5339,
|
|
"mean_token_accuracy": 0.1458686277270317,
|
|
"num_tokens": 6553390.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.582724618911743,
|
|
"epoch": 2.994452849218356,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044791963991448427,
|
|
"loss": 5.3515,
|
|
"mean_token_accuracy": 0.1610235646367073,
|
|
"num_tokens": 6563754.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.524975347518921,
|
|
"epoch": 2.999495713565305,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00044766557995713915,
|
|
"loss": 5.4385,
|
|
"mean_token_accuracy": 0.1464684471487999,
|
|
"num_tokens": 6574124.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.584632873535156,
|
|
"epoch": 3.0040342914775593,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004474109833511885,
|
|
"loss": 5.2302,
|
|
"mean_token_accuracy": 0.1594168312019772,
|
|
"num_tokens": 6582795.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.487065649032592,
|
|
"epoch": 3.0090771558245084,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004471558508879244,
|
|
"loss": 5.1521,
|
|
"mean_token_accuracy": 0.17096262723207473,
|
|
"num_tokens": 6593649.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.500127458572388,
|
|
"epoch": 3.0141200201714575,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004469001833603045,
|
|
"loss": 5.1555,
|
|
"mean_token_accuracy": 0.1699725031852722,
|
|
"num_tokens": 6606370.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.523454761505127,
|
|
"epoch": 3.0191628845184066,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004466439815629493,
|
|
"loss": 5.102,
|
|
"mean_token_accuracy": 0.16770666241645812,
|
|
"num_tokens": 6616599.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.58970365524292,
|
|
"epoch": 3.0242057488653553,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044638724629214,
|
|
"loss": 5.194,
|
|
"mean_token_accuracy": 0.1603380024433136,
|
|
"num_tokens": 6627707.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.0242057488653553,
|
|
"eval_entropy": 5.464859989156553,
|
|
"eval_loss": 5.837061405181885,
|
|
"eval_mean_token_accuracy": 0.14111255103622683,
|
|
"eval_num_tokens": 6627707.0,
|
|
"eval_runtime": 2.6605,
|
|
"eval_samples_per_second": 1179.48,
|
|
"eval_steps_per_second": 147.717,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.508710479736328,
|
|
"epoch": 3.0292486132123044,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044612997834581557,
|
|
"loss": 5.1061,
|
|
"mean_token_accuracy": 0.166281121969223,
|
|
"num_tokens": 6639001.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.460995626449585,
|
|
"epoch": 3.0342914775592535,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044587217852357086,
|
|
"loss": 5.1654,
|
|
"mean_token_accuracy": 0.1642779991030693,
|
|
"num_tokens": 6649318.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.485589694976807,
|
|
"epoch": 3.0393343419062027,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044561384762665363,
|
|
"loss": 5.2145,
|
|
"mean_token_accuracy": 0.16070360392332078,
|
|
"num_tokens": 6660584.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.4776569366455075,
|
|
"epoch": 3.044377206253152,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004453549864579623,
|
|
"loss": 5.0496,
|
|
"mean_token_accuracy": 0.17387867718935013,
|
|
"num_tokens": 6671361.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.524082851409912,
|
|
"epoch": 3.049420070600101,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004450955958220433,
|
|
"loss": 5.3019,
|
|
"mean_token_accuracy": 0.15830719023942946,
|
|
"num_tokens": 6681896.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.495604610443115,
|
|
"epoch": 3.05446293494705,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00044483567652508864,
|
|
"loss": 5.1526,
|
|
"mean_token_accuracy": 0.16557196527719498,
|
|
"num_tokens": 6693360.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.50893988609314,
|
|
"epoch": 3.059505799293999,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004445752293749337,
|
|
"loss": 5.2173,
|
|
"mean_token_accuracy": 0.16287909597158431,
|
|
"num_tokens": 6704921.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.494605493545532,
|
|
"epoch": 3.0645486636409482,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004443142551810541,
|
|
"loss": 5.1528,
|
|
"mean_token_accuracy": 0.1641039714217186,
|
|
"num_tokens": 6715659.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.41555118560791,
|
|
"epoch": 3.069591527987897,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044405275475456365,
|
|
"loss": 5.0953,
|
|
"mean_token_accuracy": 0.17433864623308182,
|
|
"num_tokens": 6726431.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.4583734512329105,
|
|
"epoch": 3.074634392334846,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004437907289082116,
|
|
"loss": 5.1735,
|
|
"mean_token_accuracy": 0.16899368017911912,
|
|
"num_tokens": 6738002.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.508736610412598,
|
|
"epoch": 3.079677256681795,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044352817845638053,
|
|
"loss": 5.1267,
|
|
"mean_token_accuracy": 0.1695928692817688,
|
|
"num_tokens": 6748509.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.487685298919677,
|
|
"epoch": 3.0847201210287443,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004432651042150833,
|
|
"loss": 5.1785,
|
|
"mean_token_accuracy": 0.16454828530550003,
|
|
"num_tokens": 6759258.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.454632186889649,
|
|
"epoch": 3.0897629853756934,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044300150700196054,
|
|
"loss": 5.1982,
|
|
"mean_token_accuracy": 0.16186091303825378,
|
|
"num_tokens": 6771263.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.470084762573242,
|
|
"epoch": 3.0948058497226425,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004427373876362786,
|
|
"loss": 5.1603,
|
|
"mean_token_accuracy": 0.16727634370326996,
|
|
"num_tokens": 6779965.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.493002986907959,
|
|
"epoch": 3.0998487140695916,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004424727469389265,
|
|
"loss": 5.1377,
|
|
"mean_token_accuracy": 0.16506763100624083,
|
|
"num_tokens": 6791521.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.497962570190429,
|
|
"epoch": 3.1048915784165407,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004422075857324137,
|
|
"loss": 5.2057,
|
|
"mean_token_accuracy": 0.16588185727596283,
|
|
"num_tokens": 6803537.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.492455530166626,
|
|
"epoch": 3.10993444276349,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004419419048408673,
|
|
"loss": 5.2062,
|
|
"mean_token_accuracy": 0.16200020164251328,
|
|
"num_tokens": 6813811.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.410846853256226,
|
|
"epoch": 3.1149773071104385,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004416757050900297,
|
|
"loss": 5.1333,
|
|
"mean_token_accuracy": 0.16689205914735794,
|
|
"num_tokens": 6825422.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.604042100906372,
|
|
"epoch": 3.1200201714573876,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004414089873072557,
|
|
"loss": 5.2864,
|
|
"mean_token_accuracy": 0.1600692167878151,
|
|
"num_tokens": 6837086.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.485888719558716,
|
|
"epoch": 3.1250630358043368,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004411417523215105,
|
|
"loss": 5.129,
|
|
"mean_token_accuracy": 0.17212664037942887,
|
|
"num_tokens": 6847195.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.383255529403686,
|
|
"epoch": 3.130105900151286,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004408740009633665,
|
|
"loss": 5.1364,
|
|
"mean_token_accuracy": 0.16953466981649398,
|
|
"num_tokens": 6857492.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.480805826187134,
|
|
"epoch": 3.135148764498235,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00044060573406500117,
|
|
"loss": 5.3231,
|
|
"mean_token_accuracy": 0.15309304147958755,
|
|
"num_tokens": 6869190.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.478493595123291,
|
|
"epoch": 3.140191628845184,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004403369524601941,
|
|
"loss": 5.1222,
|
|
"mean_token_accuracy": 0.16348316818475722,
|
|
"num_tokens": 6880728.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.403803443908691,
|
|
"epoch": 3.145234493192133,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004400676569843249,
|
|
"loss": 5.1482,
|
|
"mean_token_accuracy": 0.1617557257413864,
|
|
"num_tokens": 6892043.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.464474678039551,
|
|
"epoch": 3.1502773575390823,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043979784847436994,
|
|
"loss": 5.171,
|
|
"mean_token_accuracy": 0.17526663690805436,
|
|
"num_tokens": 6903913.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.360471248626709,
|
|
"epoch": 3.1553202218860315,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004395275277689004,
|
|
"loss": 5.1559,
|
|
"mean_token_accuracy": 0.1623574376106262,
|
|
"num_tokens": 6913951.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.369834232330322,
|
|
"epoch": 3.16036308623298,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004392566957080793,
|
|
"loss": 5.1603,
|
|
"mean_token_accuracy": 0.16323016285896302,
|
|
"num_tokens": 6924762.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.4376220703125,
|
|
"epoch": 3.1654059505799292,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00043898535313365884,
|
|
"loss": 5.1318,
|
|
"mean_token_accuracy": 0.1645357847213745,
|
|
"num_tokens": 6935845.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.3934015274047855,
|
|
"epoch": 3.1704488149268784,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000438713500888978,
|
|
"loss": 5.1835,
|
|
"mean_token_accuracy": 0.1633390322327614,
|
|
"num_tokens": 6947338.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.484806156158447,
|
|
"epoch": 3.1754916792738275,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004384411398189599,
|
|
"loss": 5.2055,
|
|
"mean_token_accuracy": 0.16281524300575256,
|
|
"num_tokens": 6958084.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.450452852249145,
|
|
"epoch": 3.1805345436207766,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000438168270770109,
|
|
"loss": 5.1426,
|
|
"mean_token_accuracy": 0.16840344965457915,
|
|
"num_tokens": 6968063.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.449947214126587,
|
|
"epoch": 3.1855774079677257,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004378948945905085,
|
|
"loss": 5.1409,
|
|
"mean_token_accuracy": 0.1703096568584442,
|
|
"num_tokens": 6978424.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.359123659133911,
|
|
"epoch": 3.190620272314675,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00043762101212981805,
|
|
"loss": 5.141,
|
|
"mean_token_accuracy": 0.16534530520439147,
|
|
"num_tokens": 6991067.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.39420714378357,
|
|
"epoch": 3.195663136661624,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043734662423927054,
|
|
"loss": 5.2196,
|
|
"mean_token_accuracy": 0.16757993549108505,
|
|
"num_tokens": 7000642.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.552215051651001,
|
|
"epoch": 3.200706001008573,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004370717317716699,
|
|
"loss": 5.2428,
|
|
"mean_token_accuracy": 0.16152413934469223,
|
|
"num_tokens": 7011305.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.489600563049317,
|
|
"epoch": 3.205748865355522,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004367963355813882,
|
|
"loss": 5.1791,
|
|
"mean_token_accuracy": 0.16151108741760253,
|
|
"num_tokens": 7021452.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.401028251647949,
|
|
"epoch": 3.210791729702471,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043652043652436323,
|
|
"loss": 5.2607,
|
|
"mean_token_accuracy": 0.1565679907798767,
|
|
"num_tokens": 7033189.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.513680505752563,
|
|
"epoch": 3.21583459404942,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004362440354580956,
|
|
"loss": 5.2041,
|
|
"mean_token_accuracy": 0.16091608256101608,
|
|
"num_tokens": 7044153.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.465910291671753,
|
|
"epoch": 3.220877458396369,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004359671332416462,
|
|
"loss": 5.267,
|
|
"mean_token_accuracy": 0.15532111674547194,
|
|
"num_tokens": 7056679.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.456948900222779,
|
|
"epoch": 3.225920322743318,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004356897307356335,
|
|
"loss": 5.1268,
|
|
"mean_token_accuracy": 0.167019584774971,
|
|
"num_tokens": 7067428.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.431487035751343,
|
|
"epoch": 3.2309631870902673,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004354118288022309,
|
|
"loss": 5.2424,
|
|
"mean_token_accuracy": 0.15960654467344285,
|
|
"num_tokens": 7079196.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.452811622619629,
|
|
"epoch": 3.2360060514372164,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043513342830516405,
|
|
"loss": 5.128,
|
|
"mean_token_accuracy": 0.16414072811603547,
|
|
"num_tokens": 7088936.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.419294929504394,
|
|
"epoch": 3.2410489157841655,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000434854530109708,
|
|
"loss": 5.2432,
|
|
"mean_token_accuracy": 0.16197886019945146,
|
|
"num_tokens": 7101504.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.498925828933716,
|
|
"epoch": 3.2460917801311147,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00043457513508268487,
|
|
"loss": 5.2075,
|
|
"mean_token_accuracy": 0.16442779302597046,
|
|
"num_tokens": 7114121.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.336520099639893,
|
|
"epoch": 3.2511346444780633,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00043429524409246083,
|
|
"loss": 5.1155,
|
|
"mean_token_accuracy": 0.16350860744714737,
|
|
"num_tokens": 7123321.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.420355033874512,
|
|
"epoch": 3.2561775088250124,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004340148580089436,
|
|
"loss": 5.1863,
|
|
"mean_token_accuracy": 0.16267697066068648,
|
|
"num_tokens": 7134145.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.557558917999268,
|
|
"epoch": 3.2612203731719616,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004337339777035794,
|
|
"loss": 5.2407,
|
|
"mean_token_accuracy": 0.1607226997613907,
|
|
"num_tokens": 7144727.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.338299369812011,
|
|
"epoch": 3.2662632375189107,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000433452604049351,
|
|
"loss": 5.1327,
|
|
"mean_token_accuracy": 0.17556441724300384,
|
|
"num_tokens": 7155531.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.39962158203125,
|
|
"epoch": 3.27130610186586,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00043317073792077394,
|
|
"loss": 5.1737,
|
|
"mean_token_accuracy": 0.16877417266368866,
|
|
"num_tokens": 7167234.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.474264240264892,
|
|
"epoch": 3.276348966212809,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004328883801938947,
|
|
"loss": 5.2552,
|
|
"mean_token_accuracy": 0.15552335381507873,
|
|
"num_tokens": 7178556.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.365787124633789,
|
|
"epoch": 3.281391830559758,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043260553174628776,
|
|
"loss": 5.1134,
|
|
"mean_token_accuracy": 0.1659499853849411,
|
|
"num_tokens": 7189074.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.473670148849488,
|
|
"epoch": 3.286434694906707,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004323221934570524,
|
|
"loss": 5.2605,
|
|
"mean_token_accuracy": 0.16382382214069366,
|
|
"num_tokens": 7200567.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.479973316192627,
|
|
"epoch": 3.2914775592536563,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00043203836620681074,
|
|
"loss": 5.2929,
|
|
"mean_token_accuracy": 0.15617599040269853,
|
|
"num_tokens": 7211012.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.408528900146484,
|
|
"epoch": 3.2965204236006054,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00043175405087770435,
|
|
"loss": 5.1925,
|
|
"mean_token_accuracy": 0.16841529905796052,
|
|
"num_tokens": 7222271.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.414000082015991,
|
|
"epoch": 3.301563287947554,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00043146924835339186,
|
|
"loss": 5.1375,
|
|
"mean_token_accuracy": 0.16959647685289383,
|
|
"num_tokens": 7235067.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.410538673400879,
|
|
"epoch": 3.306606152294503,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043118395951904605,
|
|
"loss": 5.2133,
|
|
"mean_token_accuracy": 0.16099776178598404,
|
|
"num_tokens": 7246159.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.528522968292236,
|
|
"epoch": 3.3116490166414523,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004308981852613512,
|
|
"loss": 5.2489,
|
|
"mean_token_accuracy": 0.15983354300260544,
|
|
"num_tokens": 7255944.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.461428928375244,
|
|
"epoch": 3.3166918809884014,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004306119264685003,
|
|
"loss": 5.1784,
|
|
"mean_token_accuracy": 0.16542329639196396,
|
|
"num_tokens": 7267273.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.382791900634766,
|
|
"epoch": 3.3217347453353505,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004303251840301923,
|
|
"loss": 5.1822,
|
|
"mean_token_accuracy": 0.16305022090673446,
|
|
"num_tokens": 7277622.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.427071475982666,
|
|
"epoch": 3.3267776096822996,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00043003795883762933,
|
|
"loss": 5.2207,
|
|
"mean_token_accuracy": 0.1590432867407799,
|
|
"num_tokens": 7288851.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.549104166030884,
|
|
"epoch": 3.3318204740292487,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00042975025178351386,
|
|
"loss": 5.2771,
|
|
"mean_token_accuracy": 0.15939899832010268,
|
|
"num_tokens": 7302175.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.412518739700317,
|
|
"epoch": 3.336863338376198,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000429462063762046,
|
|
"loss": 5.1436,
|
|
"mean_token_accuracy": 0.17406332194805146,
|
|
"num_tokens": 7312693.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.383169078826905,
|
|
"epoch": 3.3419062027231465,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042917339566892095,
|
|
"loss": 5.1945,
|
|
"mean_token_accuracy": 0.15786968022584916,
|
|
"num_tokens": 7323790.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.3746013164520265,
|
|
"epoch": 3.3469490670700957,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004288842484013257,
|
|
"loss": 5.1682,
|
|
"mean_token_accuracy": 0.16247669309377671,
|
|
"num_tokens": 7333797.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.412357234954834,
|
|
"epoch": 3.3519919314170448,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004285946228579365,
|
|
"loss": 5.2045,
|
|
"mean_token_accuracy": 0.17014975994825363,
|
|
"num_tokens": 7344674.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.426461029052734,
|
|
"epoch": 3.357034795763994,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004283045199389164,
|
|
"loss": 5.2039,
|
|
"mean_token_accuracy": 0.16485295295715333,
|
|
"num_tokens": 7355978.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.539542484283447,
|
|
"epoch": 3.362077660110943,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042801394054591197,
|
|
"loss": 5.305,
|
|
"mean_token_accuracy": 0.15951313972473144,
|
|
"num_tokens": 7367054.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.476680278778076,
|
|
"epoch": 3.367120524457892,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00042772288558205057,
|
|
"loss": 5.2612,
|
|
"mean_token_accuracy": 0.15854859352111816,
|
|
"num_tokens": 7379588.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.364924192428589,
|
|
"epoch": 3.3721633888048412,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004274313559519377,
|
|
"loss": 5.2333,
|
|
"mean_token_accuracy": 0.16314974874258042,
|
|
"num_tokens": 7392411.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.423452281951905,
|
|
"epoch": 3.3772062531517903,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004271393525616544,
|
|
"loss": 5.1184,
|
|
"mean_token_accuracy": 0.164414843916893,
|
|
"num_tokens": 7402801.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.406583690643311,
|
|
"epoch": 3.3822491174987395,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00042684687631875377,
|
|
"loss": 5.1875,
|
|
"mean_token_accuracy": 0.15710319727659225,
|
|
"num_tokens": 7413084.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.37700080871582,
|
|
"epoch": 3.3872919818456886,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00042655392813225864,
|
|
"loss": 5.1344,
|
|
"mean_token_accuracy": 0.16645344942808152,
|
|
"num_tokens": 7424004.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.320040845870972,
|
|
"epoch": 3.3923348461926373,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042626050891265884,
|
|
"loss": 5.1608,
|
|
"mean_token_accuracy": 0.17361150085926055,
|
|
"num_tokens": 7434461.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.450391435623169,
|
|
"epoch": 3.3973777105395864,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004259666195719081,
|
|
"loss": 5.21,
|
|
"mean_token_accuracy": 0.16076228618621827,
|
|
"num_tokens": 7445552.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.460859870910644,
|
|
"epoch": 3.4024205748865355,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042567226102342134,
|
|
"loss": 5.2055,
|
|
"mean_token_accuracy": 0.1618820071220398,
|
|
"num_tokens": 7457073.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.386827087402343,
|
|
"epoch": 3.4074634392334846,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004253774341820715,
|
|
"loss": 5.2015,
|
|
"mean_token_accuracy": 0.1622549772262573,
|
|
"num_tokens": 7468197.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.428202962875366,
|
|
"epoch": 3.4125063035804337,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004250821399641874,
|
|
"loss": 5.158,
|
|
"mean_token_accuracy": 0.16729331761598587,
|
|
"num_tokens": 7480751.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.393211221694946,
|
|
"epoch": 3.417549167927383,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004247863792875502,
|
|
"loss": 5.1787,
|
|
"mean_token_accuracy": 0.17017877399921416,
|
|
"num_tokens": 7492259.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.396057796478272,
|
|
"epoch": 3.422592032274332,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042449015307139076,
|
|
"loss": 5.2716,
|
|
"mean_token_accuracy": 0.15931643843650817,
|
|
"num_tokens": 7504317.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.412095546722412,
|
|
"epoch": 3.427634896621281,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042419346223638725,
|
|
"loss": 5.1827,
|
|
"mean_token_accuracy": 0.15693859606981278,
|
|
"num_tokens": 7516039.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.469624376296997,
|
|
"epoch": 3.4326777609682297,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004238963077046616,
|
|
"loss": 5.2225,
|
|
"mean_token_accuracy": 0.16526761502027512,
|
|
"num_tokens": 7526095.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.359675693511963,
|
|
"epoch": 3.437720625315179,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004235986903997767,
|
|
"loss": 5.1686,
|
|
"mean_token_accuracy": 0.1636862725019455,
|
|
"num_tokens": 7537320.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.419049978256226,
|
|
"epoch": 3.442763489662128,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042330061124673437,
|
|
"loss": 5.2412,
|
|
"mean_token_accuracy": 0.16660376638174057,
|
|
"num_tokens": 7548100.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.444922924041748,
|
|
"epoch": 3.447806354009077,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042300207117197116,
|
|
"loss": 5.2556,
|
|
"mean_token_accuracy": 0.15782784521579743,
|
|
"num_tokens": 7560483.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.43258376121521,
|
|
"epoch": 3.452849218356026,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042270307110335657,
|
|
"loss": 5.1751,
|
|
"mean_token_accuracy": 0.16671013683080674,
|
|
"num_tokens": 7571256.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.433862590789795,
|
|
"epoch": 3.4578920827029753,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004224036119701899,
|
|
"loss": 5.1606,
|
|
"mean_token_accuracy": 0.1657199591398239,
|
|
"num_tokens": 7581716.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.387713861465454,
|
|
"epoch": 3.4629349470499244,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042210369470319694,
|
|
"loss": 5.187,
|
|
"mean_token_accuracy": 0.17281804531812667,
|
|
"num_tokens": 7591932.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.377019357681275,
|
|
"epoch": 3.4679778113968736,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004218033202345274,
|
|
"loss": 5.1854,
|
|
"mean_token_accuracy": 0.16309025287628173,
|
|
"num_tokens": 7602842.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.379820871353149,
|
|
"epoch": 3.4730206757438227,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004215024894977522,
|
|
"loss": 5.1557,
|
|
"mean_token_accuracy": 0.1622964009642601,
|
|
"num_tokens": 7614075.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.48748631477356,
|
|
"epoch": 3.478063540090772,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004212012034278601,
|
|
"loss": 5.3368,
|
|
"mean_token_accuracy": 0.15681455731391908,
|
|
"num_tokens": 7625642.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.437462520599365,
|
|
"epoch": 3.4831064044377205,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042089946296125513,
|
|
"loss": 5.1879,
|
|
"mean_token_accuracy": 0.16691929697990418,
|
|
"num_tokens": 7636447.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.341785097122193,
|
|
"epoch": 3.4881492687846696,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00042059726903575373,
|
|
"loss": 5.1152,
|
|
"mean_token_accuracy": 0.16832868605852128,
|
|
"num_tokens": 7648684.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.3743284225463865,
|
|
"epoch": 3.4931921331316187,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004202946225905815,
|
|
"loss": 5.1791,
|
|
"mean_token_accuracy": 0.16549742221832275,
|
|
"num_tokens": 7660864.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.370916175842285,
|
|
"epoch": 3.498234997478568,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004199915245663706,
|
|
"loss": 5.1951,
|
|
"mean_token_accuracy": 0.17021624594926835,
|
|
"num_tokens": 7671939.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.557950401306153,
|
|
"epoch": 3.503277861825517,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041968797590515673,
|
|
"loss": 5.2232,
|
|
"mean_token_accuracy": 0.1650526136159897,
|
|
"num_tokens": 7681536.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.446653127670288,
|
|
"epoch": 3.508320726172466,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004193839775503761,
|
|
"loss": 5.174,
|
|
"mean_token_accuracy": 0.1683890178799629,
|
|
"num_tokens": 7693674.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.334004592895508,
|
|
"epoch": 3.513363590519415,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041907953044686253,
|
|
"loss": 5.1821,
|
|
"mean_token_accuracy": 0.171914803981781,
|
|
"num_tokens": 7704533.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.400468206405639,
|
|
"epoch": 3.518406454866364,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004187746355408448,
|
|
"loss": 5.2516,
|
|
"mean_token_accuracy": 0.16183393746614455,
|
|
"num_tokens": 7715448.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.447520923614502,
|
|
"epoch": 3.523449319213313,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004184692937799431,
|
|
"loss": 5.226,
|
|
"mean_token_accuracy": 0.16580078452825547,
|
|
"num_tokens": 7726808.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.528063488006592,
|
|
"epoch": 3.528492183560262,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00041816350611316693,
|
|
"loss": 5.2179,
|
|
"mean_token_accuracy": 0.15973225831985474,
|
|
"num_tokens": 7737899.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.528492183560262,
|
|
"eval_entropy": 5.31953394746659,
|
|
"eval_loss": 5.8183393478393555,
|
|
"eval_mean_token_accuracy": 0.14342421077134956,
|
|
"eval_num_tokens": 7737899.0,
|
|
"eval_runtime": 2.6595,
|
|
"eval_samples_per_second": 1179.915,
|
|
"eval_steps_per_second": 147.771,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.38258957862854,
|
|
"epoch": 3.533535047907211,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00041785727349091117,
|
|
"loss": 5.1704,
|
|
"mean_token_accuracy": 0.1693768784403801,
|
|
"num_tokens": 7749268.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.480043125152588,
|
|
"epoch": 3.5385779122541603,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000417550596864954,
|
|
"loss": 5.2456,
|
|
"mean_token_accuracy": 0.15714906007051468,
|
|
"num_tokens": 7760312.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.391533374786377,
|
|
"epoch": 3.5436207766011094,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00041724347718845344,
|
|
"loss": 5.2332,
|
|
"mean_token_accuracy": 0.16302382946014404,
|
|
"num_tokens": 7771130.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.428888368606567,
|
|
"epoch": 3.5486636409480585,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00041693591541594455,
|
|
"loss": 5.188,
|
|
"mean_token_accuracy": 0.16282152384519577,
|
|
"num_tokens": 7783068.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.389735555648803,
|
|
"epoch": 3.5537065052950076,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004166279125033363,
|
|
"loss": 5.2981,
|
|
"mean_token_accuracy": 0.16435786336660385,
|
|
"num_tokens": 7794314.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.485348653793335,
|
|
"epoch": 3.5587493696419568,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041631946940790893,
|
|
"loss": 5.2257,
|
|
"mean_token_accuracy": 0.16686471104621886,
|
|
"num_tokens": 7806888.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.379432487487793,
|
|
"epoch": 3.563792233988906,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041601058708831077,
|
|
"loss": 5.2462,
|
|
"mean_token_accuracy": 0.1627429559826851,
|
|
"num_tokens": 7818690.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.4259504795074465,
|
|
"epoch": 3.568835098335855,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000415701266504555,
|
|
"loss": 5.1459,
|
|
"mean_token_accuracy": 0.16211986988782884,
|
|
"num_tokens": 7829321.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.413945055007934,
|
|
"epoch": 3.5738779626828037,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004153915086180173,
|
|
"loss": 5.2248,
|
|
"mean_token_accuracy": 0.16615401655435563,
|
|
"num_tokens": 7840223.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.374924945831299,
|
|
"epoch": 3.578920827029753,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004150813143914323,
|
|
"loss": 5.2101,
|
|
"mean_token_accuracy": 0.15948547869920732,
|
|
"num_tokens": 7850999.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.4525243759155275,
|
|
"epoch": 3.583963691376702,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004147706847888908,
|
|
"loss": 5.2686,
|
|
"mean_token_accuracy": 0.16123423129320144,
|
|
"num_tokens": 7861935.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.403988075256348,
|
|
"epoch": 3.589006555723651,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004144596207758367,
|
|
"loss": 5.2553,
|
|
"mean_token_accuracy": 0.16250505894422532,
|
|
"num_tokens": 7872960.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.451062202453613,
|
|
"epoch": 3.5940494200706,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004141481233190642,
|
|
"loss": 5.286,
|
|
"mean_token_accuracy": 0.1611800268292427,
|
|
"num_tokens": 7883881.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.450012302398681,
|
|
"epoch": 3.5990922844175492,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004138361933867145,
|
|
"loss": 5.285,
|
|
"mean_token_accuracy": 0.15788197219371797,
|
|
"num_tokens": 7895106.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.410325574874878,
|
|
"epoch": 3.6041351487644984,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004135238319482731,
|
|
"loss": 5.1929,
|
|
"mean_token_accuracy": 0.16700586378574372,
|
|
"num_tokens": 7906020.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.406979084014893,
|
|
"epoch": 3.609178013111447,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004132110399745666,
|
|
"loss": 5.2549,
|
|
"mean_token_accuracy": 0.1617822080850601,
|
|
"num_tokens": 7918959.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.458474922180176,
|
|
"epoch": 3.614220877458396,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004128978184377596,
|
|
"loss": 5.198,
|
|
"mean_token_accuracy": 0.1634347215294838,
|
|
"num_tokens": 7929804.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.400113201141357,
|
|
"epoch": 3.6192637418053453,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004125841683113518,
|
|
"loss": 5.2294,
|
|
"mean_token_accuracy": 0.15927089303731917,
|
|
"num_tokens": 7941558.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.393302059173584,
|
|
"epoch": 3.6243066061522944,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00041227009057017523,
|
|
"loss": 5.2536,
|
|
"mean_token_accuracy": 0.1648019880056381,
|
|
"num_tokens": 7951737.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.3694041728973385,
|
|
"epoch": 3.6293494704992435,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004119555861903906,
|
|
"loss": 5.1625,
|
|
"mean_token_accuracy": 0.17559664696455002,
|
|
"num_tokens": 7963164.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.437113618850708,
|
|
"epoch": 3.6343923348461926,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004116406561494847,
|
|
"loss": 5.222,
|
|
"mean_token_accuracy": 0.16308198720216752,
|
|
"num_tokens": 7973275.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.432148408889771,
|
|
"epoch": 3.6394351991931417,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00041132530142626756,
|
|
"loss": 5.2004,
|
|
"mean_token_accuracy": 0.17112933546304704,
|
|
"num_tokens": 7984568.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.390055084228516,
|
|
"epoch": 3.644478063540091,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041100952300086884,
|
|
"loss": 5.2112,
|
|
"mean_token_accuracy": 0.15861116200685502,
|
|
"num_tokens": 7994263.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.458655786514282,
|
|
"epoch": 3.64952092788704,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00041069332185473514,
|
|
"loss": 5.1976,
|
|
"mean_token_accuracy": 0.16689541190862656,
|
|
"num_tokens": 8004881.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.364749813079834,
|
|
"epoch": 3.654563792233989,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000410376698970627,
|
|
"loss": 5.1777,
|
|
"mean_token_accuracy": 0.16689769178628922,
|
|
"num_tokens": 8015955.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.375132656097412,
|
|
"epoch": 3.659606656580938,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041005965533261554,
|
|
"loss": 5.1938,
|
|
"mean_token_accuracy": 0.1668706476688385,
|
|
"num_tokens": 8027010.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.383240365982056,
|
|
"epoch": 3.6646495209278873,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040974219192607973,
|
|
"loss": 5.1583,
|
|
"mean_token_accuracy": 0.16497441828250886,
|
|
"num_tokens": 8038564.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.356099557876587,
|
|
"epoch": 3.669692385274836,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00040942430973770323,
|
|
"loss": 5.2235,
|
|
"mean_token_accuracy": 0.16254878789186478,
|
|
"num_tokens": 8049975.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.476030588150024,
|
|
"epoch": 3.674735249621785,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004091060097554712,
|
|
"loss": 5.2196,
|
|
"mean_token_accuracy": 0.17048997879028321,
|
|
"num_tokens": 8061042.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.3977062702178955,
|
|
"epoch": 3.679778113968734,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004087872929686673,
|
|
"loss": 5.1692,
|
|
"mean_token_accuracy": 0.16930532604455947,
|
|
"num_tokens": 8071777.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.297494745254516,
|
|
"epoch": 3.6848209783156833,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00040846816036787056,
|
|
"loss": 5.2462,
|
|
"mean_token_accuracy": 0.15906448662281036,
|
|
"num_tokens": 8082387.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.460820484161377,
|
|
"epoch": 3.6898638426626325,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004081486129449527,
|
|
"loss": 5.1648,
|
|
"mean_token_accuracy": 0.17287870943546296,
|
|
"num_tokens": 8091818.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.48548059463501,
|
|
"epoch": 3.6949067070095816,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00040782865169307423,
|
|
"loss": 5.2448,
|
|
"mean_token_accuracy": 0.15980491638183594,
|
|
"num_tokens": 8103377.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.375945425033569,
|
|
"epoch": 3.6999495713565302,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004075082776066822,
|
|
"loss": 5.2575,
|
|
"mean_token_accuracy": 0.1602715089917183,
|
|
"num_tokens": 8114723.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.316479730606079,
|
|
"epoch": 3.7049924357034794,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004071874916815068,
|
|
"loss": 5.1034,
|
|
"mean_token_accuracy": 0.16916895061731338,
|
|
"num_tokens": 8124530.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.360852289199829,
|
|
"epoch": 3.7100353000504285,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00040686629491455787,
|
|
"loss": 5.1991,
|
|
"mean_token_accuracy": 0.16713961809873581,
|
|
"num_tokens": 8134894.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.35507984161377,
|
|
"epoch": 3.7150781643973776,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040654468830412247,
|
|
"loss": 5.147,
|
|
"mean_token_accuracy": 0.17256019711494447,
|
|
"num_tokens": 8143499.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.38199815750122,
|
|
"epoch": 3.7201210287443267,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00040622267284976134,
|
|
"loss": 5.1969,
|
|
"mean_token_accuracy": 0.16475919634103775,
|
|
"num_tokens": 8154556.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.380331420898438,
|
|
"epoch": 3.725163893091276,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00040590024955230596,
|
|
"loss": 5.3141,
|
|
"mean_token_accuracy": 0.15468263775110244,
|
|
"num_tokens": 8166047.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.441823148727417,
|
|
"epoch": 3.730206757438225,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004055774194138554,
|
|
"loss": 5.2038,
|
|
"mean_token_accuracy": 0.1651650220155716,
|
|
"num_tokens": 8177068.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.389064073562622,
|
|
"epoch": 3.735249621785174,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004052541834377731,
|
|
"loss": 5.1938,
|
|
"mean_token_accuracy": 0.16553791463375092,
|
|
"num_tokens": 8187316.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.405516052246094,
|
|
"epoch": 3.740292486132123,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00040493054262868397,
|
|
"loss": 5.3062,
|
|
"mean_token_accuracy": 0.1629395678639412,
|
|
"num_tokens": 8198642.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.382053422927856,
|
|
"epoch": 3.7453353504790723,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040460649799247116,
|
|
"loss": 5.1636,
|
|
"mean_token_accuracy": 0.1711382031440735,
|
|
"num_tokens": 8208476.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.356726932525635,
|
|
"epoch": 3.7503782148260214,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004042820505362728,
|
|
"loss": 5.2512,
|
|
"mean_token_accuracy": 0.1641302540898323,
|
|
"num_tokens": 8219590.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.367624759674072,
|
|
"epoch": 3.7554210791729705,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004039572012684792,
|
|
"loss": 5.1449,
|
|
"mean_token_accuracy": 0.17125097662210464,
|
|
"num_tokens": 8230707.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.456900930404663,
|
|
"epoch": 3.760463943519919,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00040363195119872916,
|
|
"loss": 5.2127,
|
|
"mean_token_accuracy": 0.15754593908786774,
|
|
"num_tokens": 8241774.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.503872299194336,
|
|
"epoch": 3.7655068078668683,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00040330630133790757,
|
|
"loss": 5.3445,
|
|
"mean_token_accuracy": 0.15596480369567872,
|
|
"num_tokens": 8254314.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.424444341659546,
|
|
"epoch": 3.7705496722138174,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040298025269814165,
|
|
"loss": 5.1771,
|
|
"mean_token_accuracy": 0.16738559305667877,
|
|
"num_tokens": 8265650.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.355651140213013,
|
|
"epoch": 3.7755925365607665,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004026538062927981,
|
|
"loss": 5.1507,
|
|
"mean_token_accuracy": 0.16871436834335327,
|
|
"num_tokens": 8276808.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.348376035690308,
|
|
"epoch": 3.7806354009077157,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00040232696313647985,
|
|
"loss": 5.0976,
|
|
"mean_token_accuracy": 0.1650240823626518,
|
|
"num_tokens": 8287867.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.392797946929932,
|
|
"epoch": 3.7856782652546648,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004019997242450231,
|
|
"loss": 5.1544,
|
|
"mean_token_accuracy": 0.1688045382499695,
|
|
"num_tokens": 8298707.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.302408933639526,
|
|
"epoch": 3.7907211296016134,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004016720906354937,
|
|
"loss": 5.1141,
|
|
"mean_token_accuracy": 0.17131074219942094,
|
|
"num_tokens": 8309363.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.400753211975098,
|
|
"epoch": 3.7957639939485626,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004013440633261845,
|
|
"loss": 5.2886,
|
|
"mean_token_accuracy": 0.16195887625217437,
|
|
"num_tokens": 8319376.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.398118257522583,
|
|
"epoch": 3.8008068582955117,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00040101564333661213,
|
|
"loss": 5.1774,
|
|
"mean_token_accuracy": 0.1709526926279068,
|
|
"num_tokens": 8330842.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.320674228668213,
|
|
"epoch": 3.805849722642461,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00040068683168751316,
|
|
"loss": 5.1409,
|
|
"mean_token_accuracy": 0.16858583241701125,
|
|
"num_tokens": 8341588.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.364935827255249,
|
|
"epoch": 3.81089258698941,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00040035762940084205,
|
|
"loss": 5.1884,
|
|
"mean_token_accuracy": 0.1611656665802002,
|
|
"num_tokens": 8351664.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.484912300109864,
|
|
"epoch": 3.815935451336359,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000400028037499767,
|
|
"loss": 5.2762,
|
|
"mean_token_accuracy": 0.15862194299697877,
|
|
"num_tokens": 8362803.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.331174850463867,
|
|
"epoch": 3.820978315683308,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003996980570086673,
|
|
"loss": 5.2091,
|
|
"mean_token_accuracy": 0.16397041827440262,
|
|
"num_tokens": 8373791.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.405066776275635,
|
|
"epoch": 3.8260211800302573,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039936768895312987,
|
|
"loss": 5.1972,
|
|
"mean_token_accuracy": 0.16669903546571732,
|
|
"num_tokens": 8385738.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.372370815277099,
|
|
"epoch": 3.8310640443772064,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00039903693435994626,
|
|
"loss": 5.2084,
|
|
"mean_token_accuracy": 0.16176739037036897,
|
|
"num_tokens": 8396924.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.428881549835205,
|
|
"epoch": 3.8361069087241555,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00039870579425710937,
|
|
"loss": 5.2898,
|
|
"mean_token_accuracy": 0.15452689081430435,
|
|
"num_tokens": 8409040.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.317065811157226,
|
|
"epoch": 3.8411497730711046,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003983742696738103,
|
|
"loss": 5.1279,
|
|
"mean_token_accuracy": 0.16751042604446412,
|
|
"num_tokens": 8419630.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.434832334518433,
|
|
"epoch": 3.8461926374180537,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003980423616404351,
|
|
"loss": 5.2699,
|
|
"mean_token_accuracy": 0.16454644054174422,
|
|
"num_tokens": 8430036.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.408985757827759,
|
|
"epoch": 3.8512355017650024,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003977100711885616,
|
|
"loss": 5.1398,
|
|
"mean_token_accuracy": 0.17262270897626877,
|
|
"num_tokens": 8439829.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.333293485641479,
|
|
"epoch": 3.8562783661119515,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00039737739935095614,
|
|
"loss": 5.1705,
|
|
"mean_token_accuracy": 0.1661990240216255,
|
|
"num_tokens": 8450492.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.339181566238404,
|
|
"epoch": 3.8613212304589006,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003970443471615705,
|
|
"loss": 5.2469,
|
|
"mean_token_accuracy": 0.1617508724331856,
|
|
"num_tokens": 8462200.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.388348388671875,
|
|
"epoch": 3.8663640948058497,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003967109156555386,
|
|
"loss": 5.1938,
|
|
"mean_token_accuracy": 0.1618059456348419,
|
|
"num_tokens": 8473393.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.368866062164306,
|
|
"epoch": 3.871406959152799,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039637710586917324,
|
|
"loss": 5.1759,
|
|
"mean_token_accuracy": 0.16557400971651076,
|
|
"num_tokens": 8484646.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.415713691711426,
|
|
"epoch": 3.876449823499748,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00039604291883996287,
|
|
"loss": 5.2169,
|
|
"mean_token_accuracy": 0.16508596986532212,
|
|
"num_tokens": 8495371.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.363277339935303,
|
|
"epoch": 3.8814926878466967,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039570835560656854,
|
|
"loss": 5.2167,
|
|
"mean_token_accuracy": 0.16425310373306273,
|
|
"num_tokens": 8507976.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.381908798217774,
|
|
"epoch": 3.8865355521936458,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003953734172088205,
|
|
"loss": 5.2269,
|
|
"mean_token_accuracy": 0.16618974208831788,
|
|
"num_tokens": 8519120.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.369919776916504,
|
|
"epoch": 3.891578416540595,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039503810468771494,
|
|
"loss": 5.2038,
|
|
"mean_token_accuracy": 0.16709574013948442,
|
|
"num_tokens": 8529727.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.360252857208252,
|
|
"epoch": 3.896621280887544,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003947024190854111,
|
|
"loss": 5.1872,
|
|
"mean_token_accuracy": 0.1645798221230507,
|
|
"num_tokens": 8540202.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.442239284515381,
|
|
"epoch": 3.901664145234493,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039436636144522737,
|
|
"loss": 5.1651,
|
|
"mean_token_accuracy": 0.17157187908887864,
|
|
"num_tokens": 8550810.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.320820379257202,
|
|
"epoch": 3.9067070095814422,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00039402993281163875,
|
|
"loss": 5.1783,
|
|
"mean_token_accuracy": 0.1672923371195793,
|
|
"num_tokens": 8561120.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.356241846084595,
|
|
"epoch": 3.9117498739283914,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003936931342302732,
|
|
"loss": 5.2156,
|
|
"mean_token_accuracy": 0.1667876958847046,
|
|
"num_tokens": 8572153.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.398838853836059,
|
|
"epoch": 3.9167927382753405,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003933559667479084,
|
|
"loss": 5.2007,
|
|
"mean_token_accuracy": 0.16505681723356247,
|
|
"num_tokens": 8583740.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.375427293777466,
|
|
"epoch": 3.9218356026222896,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039301843141246876,
|
|
"loss": 5.2275,
|
|
"mean_token_accuracy": 0.16356826275587083,
|
|
"num_tokens": 8596017.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.421029663085937,
|
|
"epoch": 3.9268784669692387,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039268052927302185,
|
|
"loss": 5.2202,
|
|
"mean_token_accuracy": 0.162451808154583,
|
|
"num_tokens": 8608121.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.397237253189087,
|
|
"epoch": 3.931921331316188,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003923422613797753,
|
|
"loss": 5.2376,
|
|
"mean_token_accuracy": 0.16345153003931046,
|
|
"num_tokens": 8621446.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.40749225616455,
|
|
"epoch": 3.936964195663137,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003920036287840734,
|
|
"loss": 5.2332,
|
|
"mean_token_accuracy": 0.16792467832565308,
|
|
"num_tokens": 8632088.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.335327577590943,
|
|
"epoch": 3.9420070600100856,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039166463253839434,
|
|
"loss": 5.0937,
|
|
"mean_token_accuracy": 0.17464697360992432,
|
|
"num_tokens": 8642546.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.354379940032959,
|
|
"epoch": 3.9470499243570347,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003913252736963462,
|
|
"loss": 5.2052,
|
|
"mean_token_accuracy": 0.16449425667524337,
|
|
"num_tokens": 8652869.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.426086139678955,
|
|
"epoch": 3.952092788703984,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000390985553312664,
|
|
"loss": 5.1432,
|
|
"mean_token_accuracy": 0.16367359161376954,
|
|
"num_tokens": 8662834.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.404352855682373,
|
|
"epoch": 3.957135653050933,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003906454724432065,
|
|
"loss": 5.2753,
|
|
"mean_token_accuracy": 0.1602453410625458,
|
|
"num_tokens": 8673372.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.365799474716186,
|
|
"epoch": 3.962178517397882,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039030503214495296,
|
|
"loss": 5.1913,
|
|
"mean_token_accuracy": 0.16619714498519897,
|
|
"num_tokens": 8683660.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.410293674468994,
|
|
"epoch": 3.967221381744831,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003899642334759996,
|
|
"loss": 5.2263,
|
|
"mean_token_accuracy": 0.1586088165640831,
|
|
"num_tokens": 8694044.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.387966299057007,
|
|
"epoch": 3.97226424609178,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003896230774955567,
|
|
"loss": 5.2352,
|
|
"mean_token_accuracy": 0.15973966866731643,
|
|
"num_tokens": 8705014.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.395192623138428,
|
|
"epoch": 3.977307110438729,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00038928156526394476,
|
|
"loss": 5.1827,
|
|
"mean_token_accuracy": 0.1619753807783127,
|
|
"num_tokens": 8716007.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.381972885131836,
|
|
"epoch": 3.982349974785678,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003889396978425918,
|
|
"loss": 5.2146,
|
|
"mean_token_accuracy": 0.16785817593336105,
|
|
"num_tokens": 8728390.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.406080722808838,
|
|
"epoch": 3.987392839132627,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003885974762940295,
|
|
"loss": 5.1446,
|
|
"mean_token_accuracy": 0.16514059156179428,
|
|
"num_tokens": 8738806.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.369142818450928,
|
|
"epoch": 3.9924357034795763,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003882549016818905,
|
|
"loss": 5.2043,
|
|
"mean_token_accuracy": 0.16388882249593734,
|
|
"num_tokens": 8749667.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.33401460647583,
|
|
"epoch": 3.9974785678265254,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00038791197507090456,
|
|
"loss": 5.2175,
|
|
"mean_token_accuracy": 0.16347086280584336,
|
|
"num_tokens": 8761380.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.354054027133518,
|
|
"epoch": 4.00201714573878,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00038756869752689527,
|
|
"loss": 5.0411,
|
|
"mean_token_accuracy": 0.17043197982841068,
|
|
"num_tokens": 8770921.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.332054853439331,
|
|
"epoch": 4.007060010085729,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003872250701167775,
|
|
"loss": 4.8231,
|
|
"mean_token_accuracy": 0.18298488557338716,
|
|
"num_tokens": 8782706.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.354917430877686,
|
|
"epoch": 4.012102874432678,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00038688109390855297,
|
|
"loss": 4.9389,
|
|
"mean_token_accuracy": 0.18452763855457305,
|
|
"num_tokens": 8794600.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.3562620162963865,
|
|
"epoch": 4.017145738779627,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003865367699713077,
|
|
"loss": 4.8341,
|
|
"mean_token_accuracy": 0.18144126534461974,
|
|
"num_tokens": 8804149.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.288506317138672,
|
|
"epoch": 4.022188603126576,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00038619209937520836,
|
|
"loss": 4.8464,
|
|
"mean_token_accuracy": 0.18950561285018921,
|
|
"num_tokens": 8814723.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.359831666946411,
|
|
"epoch": 4.027231467473525,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00038584708319149924,
|
|
"loss": 4.9474,
|
|
"mean_token_accuracy": 0.1810444787144661,
|
|
"num_tokens": 8827258.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.33565468788147,
|
|
"epoch": 4.032274331820474,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038550172249249856,
|
|
"loss": 4.9059,
|
|
"mean_token_accuracy": 0.17861006557941436,
|
|
"num_tokens": 8837528.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 4.032274331820474,
|
|
"eval_entropy": 5.208278768845187,
|
|
"eval_loss": 5.776338577270508,
|
|
"eval_mean_token_accuracy": 0.14789634662426762,
|
|
"eval_num_tokens": 8837528.0,
|
|
"eval_runtime": 2.6564,
|
|
"eval_samples_per_second": 1181.283,
|
|
"eval_steps_per_second": 147.943,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.297657871246338,
|
|
"epoch": 4.0373171961674235,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003851560183515953,
|
|
"loss": 4.8897,
|
|
"mean_token_accuracy": 0.18151450902223587,
|
|
"num_tokens": 8848154.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.276195430755616,
|
|
"epoch": 4.042360060514373,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000384809971843246,
|
|
"loss": 4.8731,
|
|
"mean_token_accuracy": 0.18539131581783294,
|
|
"num_tokens": 8858396.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.392487525939941,
|
|
"epoch": 4.047402924861321,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000384463584042971,
|
|
"loss": 4.866,
|
|
"mean_token_accuracy": 0.1902894541621208,
|
|
"num_tokens": 8868590.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.303223133087158,
|
|
"epoch": 4.05244578920827,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00038411685602735175,
|
|
"loss": 4.8887,
|
|
"mean_token_accuracy": 0.18199685364961624,
|
|
"num_tokens": 8878955.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.2945716381073,
|
|
"epoch": 4.057488653555219,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038376978887402687,
|
|
"loss": 4.8568,
|
|
"mean_token_accuracy": 0.18815067559480667,
|
|
"num_tokens": 8889402.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.304863500595093,
|
|
"epoch": 4.062531517902168,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003834223836616891,
|
|
"loss": 4.8765,
|
|
"mean_token_accuracy": 0.186817067861557,
|
|
"num_tokens": 8901433.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.218743896484375,
|
|
"epoch": 4.067574382249117,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038307464147008176,
|
|
"loss": 4.9291,
|
|
"mean_token_accuracy": 0.1806337743997574,
|
|
"num_tokens": 8912895.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.315894317626953,
|
|
"epoch": 4.072617246596066,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003827265633799957,
|
|
"loss": 4.9261,
|
|
"mean_token_accuracy": 0.1779945284128189,
|
|
"num_tokens": 8923529.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.308766078948975,
|
|
"epoch": 4.0776601109430155,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003823781504732657,
|
|
"loss": 4.835,
|
|
"mean_token_accuracy": 0.18404545336961747,
|
|
"num_tokens": 8933279.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.330894088745117,
|
|
"epoch": 4.082702975289965,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003820294038327669,
|
|
"loss": 4.947,
|
|
"mean_token_accuracy": 0.18193530887365342,
|
|
"num_tokens": 8944396.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.234827280044556,
|
|
"epoch": 4.087745839636914,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003816803245424121,
|
|
"loss": 4.7406,
|
|
"mean_token_accuracy": 0.19521126747131348,
|
|
"num_tokens": 8955535.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.305743646621704,
|
|
"epoch": 4.092788703983863,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003813309136871478,
|
|
"loss": 4.9168,
|
|
"mean_token_accuracy": 0.17832522243261337,
|
|
"num_tokens": 8965859.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.334202671051026,
|
|
"epoch": 4.097831568330812,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038098117235295083,
|
|
"loss": 4.9613,
|
|
"mean_token_accuracy": 0.18180454075336455,
|
|
"num_tokens": 8976580.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.249305486679077,
|
|
"epoch": 4.102874432677761,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003806311016268254,
|
|
"loss": 4.9085,
|
|
"mean_token_accuracy": 0.18239586353302,
|
|
"num_tokens": 8988154.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.34602632522583,
|
|
"epoch": 4.10791729702471,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003802807025967995,
|
|
"loss": 4.9405,
|
|
"mean_token_accuracy": 0.1817002162337303,
|
|
"num_tokens": 8999758.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.286632442474366,
|
|
"epoch": 4.112960161371659,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003799299763519213,
|
|
"loss": 4.8844,
|
|
"mean_token_accuracy": 0.18062795102596282,
|
|
"num_tokens": 9010371.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.281461143493653,
|
|
"epoch": 4.118003025718608,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003795789239822561,
|
|
"loss": 4.8907,
|
|
"mean_token_accuracy": 0.18999302089214326,
|
|
"num_tokens": 9021224.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.294471406936646,
|
|
"epoch": 4.1230458900655576,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003792275465788826,
|
|
"loss": 4.9508,
|
|
"mean_token_accuracy": 0.18266406804323196,
|
|
"num_tokens": 9033569.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.295239448547363,
|
|
"epoch": 4.128088754412507,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00037887584523389013,
|
|
"loss": 4.9586,
|
|
"mean_token_accuracy": 0.17855701446533204,
|
|
"num_tokens": 9046166.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.317925882339478,
|
|
"epoch": 4.133131618759456,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00037852382104037454,
|
|
"loss": 4.9348,
|
|
"mean_token_accuracy": 0.18432571440935136,
|
|
"num_tokens": 9056210.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.296660900115967,
|
|
"epoch": 4.138174483106404,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00037817147509243497,
|
|
"loss": 4.8971,
|
|
"mean_token_accuracy": 0.18163620233535765,
|
|
"num_tokens": 9067473.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.3439067840576175,
|
|
"epoch": 4.143217347453353,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003778188084851709,
|
|
"loss": 4.9499,
|
|
"mean_token_accuracy": 0.17843894213438033,
|
|
"num_tokens": 9077744.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.263043975830078,
|
|
"epoch": 4.148260211800302,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00037746582231467816,
|
|
"loss": 4.9323,
|
|
"mean_token_accuracy": 0.17945596426725388,
|
|
"num_tokens": 9088988.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.328462409973144,
|
|
"epoch": 4.153303076147251,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003771125176780461,
|
|
"loss": 4.9723,
|
|
"mean_token_accuracy": 0.17299776077270507,
|
|
"num_tokens": 9100584.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.310266160964966,
|
|
"epoch": 4.1583459404942005,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00037675889567335345,
|
|
"loss": 4.9417,
|
|
"mean_token_accuracy": 0.17984073460102082,
|
|
"num_tokens": 9112442.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.309251308441162,
|
|
"epoch": 4.16338880484115,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00037640495739966573,
|
|
"loss": 4.9873,
|
|
"mean_token_accuracy": 0.17330888509750367,
|
|
"num_tokens": 9122903.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.392861032485962,
|
|
"epoch": 4.168431669188099,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00037605070395703124,
|
|
"loss": 4.9961,
|
|
"mean_token_accuracy": 0.17788511365652085,
|
|
"num_tokens": 9134301.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.42130217552185,
|
|
"epoch": 4.173474533535048,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00037569613644647783,
|
|
"loss": 4.9925,
|
|
"mean_token_accuracy": 0.172127428650856,
|
|
"num_tokens": 9144503.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.233960199356079,
|
|
"epoch": 4.178517397881997,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003753412559700094,
|
|
"loss": 4.8312,
|
|
"mean_token_accuracy": 0.1824622705578804,
|
|
"num_tokens": 9154173.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.307043075561523,
|
|
"epoch": 4.183560262228946,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00037498606363060285,
|
|
"loss": 4.959,
|
|
"mean_token_accuracy": 0.18360354006290436,
|
|
"num_tokens": 9163772.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.325912618637085,
|
|
"epoch": 4.188603126575895,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003746305605322041,
|
|
"loss": 4.9811,
|
|
"mean_token_accuracy": 0.17522898465394973,
|
|
"num_tokens": 9174618.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.424467897415161,
|
|
"epoch": 4.193645990922844,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00037427474777972477,
|
|
"loss": 5.0323,
|
|
"mean_token_accuracy": 0.17776864171028137,
|
|
"num_tokens": 9185199.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.329036474227905,
|
|
"epoch": 4.198688855269793,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00037391862647903943,
|
|
"loss": 4.9322,
|
|
"mean_token_accuracy": 0.1820979058742523,
|
|
"num_tokens": 9197329.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.266254520416259,
|
|
"epoch": 4.2037317196167425,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00037356219773698106,
|
|
"loss": 4.9113,
|
|
"mean_token_accuracy": 0.1795494183897972,
|
|
"num_tokens": 9207516.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.275842189788818,
|
|
"epoch": 4.208774583963692,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00037320546266133854,
|
|
"loss": 4.9854,
|
|
"mean_token_accuracy": 0.1674117550253868,
|
|
"num_tokens": 9219598.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.2778326034545895,
|
|
"epoch": 4.213817448310641,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003728484223608526,
|
|
"loss": 4.8677,
|
|
"mean_token_accuracy": 0.1809871420264244,
|
|
"num_tokens": 9229675.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.248960542678833,
|
|
"epoch": 4.21886031265759,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00037249107794521286,
|
|
"loss": 4.9198,
|
|
"mean_token_accuracy": 0.1811830997467041,
|
|
"num_tokens": 9240308.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.23493537902832,
|
|
"epoch": 4.223903177004539,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00037213343052505394,
|
|
"loss": 4.9147,
|
|
"mean_token_accuracy": 0.1765119031071663,
|
|
"num_tokens": 9250299.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.302527856826782,
|
|
"epoch": 4.228946041351487,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003717754812119523,
|
|
"loss": 4.9575,
|
|
"mean_token_accuracy": 0.1776071548461914,
|
|
"num_tokens": 9262659.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.321568965911865,
|
|
"epoch": 4.233988905698436,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003714172311184226,
|
|
"loss": 4.8982,
|
|
"mean_token_accuracy": 0.18398793488740922,
|
|
"num_tokens": 9274430.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.284254264831543,
|
|
"epoch": 4.2390317700453854,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003710586813579144,
|
|
"loss": 4.9991,
|
|
"mean_token_accuracy": 0.174013651907444,
|
|
"num_tokens": 9285484.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.263162517547608,
|
|
"epoch": 4.244074634392335,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00037069983304480875,
|
|
"loss": 4.9089,
|
|
"mean_token_accuracy": 0.1845747485756874,
|
|
"num_tokens": 9296253.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.285987424850464,
|
|
"epoch": 4.249117498739284,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003703406872944144,
|
|
"loss": 4.9419,
|
|
"mean_token_accuracy": 0.1824813097715378,
|
|
"num_tokens": 9308491.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.349198818206787,
|
|
"epoch": 4.254160363086233,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003699812452229645,
|
|
"loss": 4.9213,
|
|
"mean_token_accuracy": 0.18035571724176408,
|
|
"num_tokens": 9319722.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.262891817092895,
|
|
"epoch": 4.259203227433182,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00036962150794761354,
|
|
"loss": 4.8489,
|
|
"mean_token_accuracy": 0.18958741426467896,
|
|
"num_tokens": 9329324.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.267193222045899,
|
|
"epoch": 4.264246091780131,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003692614765864331,
|
|
"loss": 4.9571,
|
|
"mean_token_accuracy": 0.18043942004442215,
|
|
"num_tokens": 9340060.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.263688373565674,
|
|
"epoch": 4.26928895612708,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00036890115225840904,
|
|
"loss": 4.9469,
|
|
"mean_token_accuracy": 0.18326425403356553,
|
|
"num_tokens": 9351071.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.328941011428833,
|
|
"epoch": 4.274331820474029,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003685405360834376,
|
|
"loss": 5.0103,
|
|
"mean_token_accuracy": 0.17122016102075577,
|
|
"num_tokens": 9361910.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.353436756134033,
|
|
"epoch": 4.279374684820978,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00036817962918232234,
|
|
"loss": 4.9368,
|
|
"mean_token_accuracy": 0.1784886583685875,
|
|
"num_tokens": 9371964.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.301136684417725,
|
|
"epoch": 4.2844175491679275,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00036781843267677,
|
|
"loss": 4.8994,
|
|
"mean_token_accuracy": 0.18513788282871246,
|
|
"num_tokens": 9381922.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.267109346389771,
|
|
"epoch": 4.289460413514877,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003674569476893878,
|
|
"loss": 4.8556,
|
|
"mean_token_accuracy": 0.1900559589266777,
|
|
"num_tokens": 9393256.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.215050125122071,
|
|
"epoch": 4.294503277861826,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003670951753436794,
|
|
"loss": 4.909,
|
|
"mean_token_accuracy": 0.18377722203731536,
|
|
"num_tokens": 9403703.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.3127134323120115,
|
|
"epoch": 4.299546142208775,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00036673311676404163,
|
|
"loss": 4.9765,
|
|
"mean_token_accuracy": 0.1857394814491272,
|
|
"num_tokens": 9414454.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.212280035018921,
|
|
"epoch": 4.304589006555724,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003663707730757608,
|
|
"loss": 4.8905,
|
|
"mean_token_accuracy": 0.18425302654504777,
|
|
"num_tokens": 9425566.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.281070423126221,
|
|
"epoch": 4.309631870902673,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00036600814540500956,
|
|
"loss": 5.03,
|
|
"mean_token_accuracy": 0.16893292367458343,
|
|
"num_tokens": 9435555.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.384689283370972,
|
|
"epoch": 4.314674735249622,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000365645234878843,
|
|
"loss": 5.0466,
|
|
"mean_token_accuracy": 0.1713443249464035,
|
|
"num_tokens": 9447784.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.321556758880615,
|
|
"epoch": 4.31971759959657,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00036528204262519545,
|
|
"loss": 4.9251,
|
|
"mean_token_accuracy": 0.18031369149684906,
|
|
"num_tokens": 9458243.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.281390237808227,
|
|
"epoch": 4.3247604639435195,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003649185697728768,
|
|
"loss": 4.9246,
|
|
"mean_token_accuracy": 0.17944859713315964,
|
|
"num_tokens": 9470258.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.301630353927612,
|
|
"epoch": 4.329803328290469,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00036455481745156906,
|
|
"loss": 4.9367,
|
|
"mean_token_accuracy": 0.18308681547641753,
|
|
"num_tokens": 9480787.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.312137174606323,
|
|
"epoch": 4.334846192637418,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003641907867918228,
|
|
"loss": 4.9186,
|
|
"mean_token_accuracy": 0.1798335686326027,
|
|
"num_tokens": 9492459.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.322690868377686,
|
|
"epoch": 4.339889056984367,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003638264789250538,
|
|
"loss": 5.0081,
|
|
"mean_token_accuracy": 0.1770971432328224,
|
|
"num_tokens": 9504600.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.382269191741943,
|
|
"epoch": 4.344931921331316,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00036346189498353914,
|
|
"loss": 5.0141,
|
|
"mean_token_accuracy": 0.17457421422004699,
|
|
"num_tokens": 9515853.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.250313806533813,
|
|
"epoch": 4.349974785678265,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003630970361004143,
|
|
"loss": 4.9957,
|
|
"mean_token_accuracy": 0.18191644698381423,
|
|
"num_tokens": 9528109.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.27856674194336,
|
|
"epoch": 4.355017650025214,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003627319034096691,
|
|
"loss": 4.9461,
|
|
"mean_token_accuracy": 0.17691521495580673,
|
|
"num_tokens": 9539795.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.284719562530517,
|
|
"epoch": 4.360060514372163,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003623664980461443,
|
|
"loss": 4.9695,
|
|
"mean_token_accuracy": 0.17641211301088333,
|
|
"num_tokens": 9551020.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.297403335571289,
|
|
"epoch": 4.3651033787191125,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00036200082114552806,
|
|
"loss": 4.9668,
|
|
"mean_token_accuracy": 0.17779829800128938,
|
|
"num_tokens": 9562540.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.3376048564910885,
|
|
"epoch": 4.370146243066062,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00036163487384435294,
|
|
"loss": 4.9993,
|
|
"mean_token_accuracy": 0.17377704232931138,
|
|
"num_tokens": 9574122.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.297678089141845,
|
|
"epoch": 4.375189107413011,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003612686572799914,
|
|
"loss": 4.9532,
|
|
"mean_token_accuracy": 0.1819907620549202,
|
|
"num_tokens": 9584781.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.363706970214844,
|
|
"epoch": 4.38023197175996,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003609021725906529,
|
|
"loss": 4.9834,
|
|
"mean_token_accuracy": 0.18166965693235398,
|
|
"num_tokens": 9595230.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.307488441467285,
|
|
"epoch": 4.385274836106909,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00036053542091538045,
|
|
"loss": 4.922,
|
|
"mean_token_accuracy": 0.18681933283805846,
|
|
"num_tokens": 9605814.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.242676687240601,
|
|
"epoch": 4.390317700453858,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00036016840339404664,
|
|
"loss": 4.9236,
|
|
"mean_token_accuracy": 0.1799074187874794,
|
|
"num_tokens": 9617900.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.247936868667603,
|
|
"epoch": 4.395360564800807,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003598011211673505,
|
|
"loss": 4.9609,
|
|
"mean_token_accuracy": 0.1835429251194,
|
|
"num_tokens": 9628987.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.292639970779419,
|
|
"epoch": 4.400403429147756,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003594335753768135,
|
|
"loss": 4.9185,
|
|
"mean_token_accuracy": 0.1842707499861717,
|
|
"num_tokens": 9638419.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.293217563629151,
|
|
"epoch": 4.405446293494705,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003590657671647766,
|
|
"loss": 4.9445,
|
|
"mean_token_accuracy": 0.17909929305315017,
|
|
"num_tokens": 9648975.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.276272535324097,
|
|
"epoch": 4.4104891578416545,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00035869769767439606,
|
|
"loss": 4.9059,
|
|
"mean_token_accuracy": 0.18617264926433563,
|
|
"num_tokens": 9659935.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.319676113128662,
|
|
"epoch": 4.415532022188603,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003583293680496404,
|
|
"loss": 5.019,
|
|
"mean_token_accuracy": 0.17957741767168045,
|
|
"num_tokens": 9671269.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.278383827209472,
|
|
"epoch": 4.420574886535552,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003579607794352867,
|
|
"loss": 4.946,
|
|
"mean_token_accuracy": 0.1803794488310814,
|
|
"num_tokens": 9682627.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.2409875869750975,
|
|
"epoch": 4.425617750882501,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00035759193297691674,
|
|
"loss": 4.8726,
|
|
"mean_token_accuracy": 0.1800372466444969,
|
|
"num_tokens": 9693482.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.2448655128479,
|
|
"epoch": 4.43066061522945,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00035722282982091373,
|
|
"loss": 4.99,
|
|
"mean_token_accuracy": 0.17834897488355636,
|
|
"num_tokens": 9705279.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.293869924545288,
|
|
"epoch": 4.435703479576399,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003568534711144591,
|
|
"loss": 4.9919,
|
|
"mean_token_accuracy": 0.17380879074335098,
|
|
"num_tokens": 9716695.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.263970041275025,
|
|
"epoch": 4.440746343923348,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000356483858005528,
|
|
"loss": 4.8949,
|
|
"mean_token_accuracy": 0.18470173478126525,
|
|
"num_tokens": 9727960.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.2939050674438475,
|
|
"epoch": 4.445789208270297,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00035611399164288654,
|
|
"loss": 4.9729,
|
|
"mean_token_accuracy": 0.17486375123262404,
|
|
"num_tokens": 9738341.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.248948860168457,
|
|
"epoch": 4.450832072617247,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003557438731760879,
|
|
"loss": 4.9801,
|
|
"mean_token_accuracy": 0.18036776930093765,
|
|
"num_tokens": 9748261.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.325256490707398,
|
|
"epoch": 4.455874936964196,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003553735037554689,
|
|
"loss": 4.9605,
|
|
"mean_token_accuracy": 0.17630374133586885,
|
|
"num_tokens": 9759109.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.314725303649903,
|
|
"epoch": 4.460917801311145,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003550028845321461,
|
|
"loss": 4.9459,
|
|
"mean_token_accuracy": 0.1791907638311386,
|
|
"num_tokens": 9770431.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.2525615215301515,
|
|
"epoch": 4.465960665658094,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003546320166580127,
|
|
"loss": 5.0021,
|
|
"mean_token_accuracy": 0.17144933789968492,
|
|
"num_tokens": 9783459.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.282162809371949,
|
|
"epoch": 4.471003530005043,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00035426090128573464,
|
|
"loss": 4.9037,
|
|
"mean_token_accuracy": 0.18521065562963485,
|
|
"num_tokens": 9793961.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.348775577545166,
|
|
"epoch": 4.476046394351992,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00035388953956874686,
|
|
"loss": 4.9917,
|
|
"mean_token_accuracy": 0.1747578874230385,
|
|
"num_tokens": 9803964.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.25225977897644,
|
|
"epoch": 4.481089258698941,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003535179326612504,
|
|
"loss": 4.9094,
|
|
"mean_token_accuracy": 0.18465292900800706,
|
|
"num_tokens": 9816176.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.2114511013031,
|
|
"epoch": 4.48613212304589,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00035314608171820787,
|
|
"loss": 4.8752,
|
|
"mean_token_accuracy": 0.18814050853252412,
|
|
"num_tokens": 9826649.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.339834785461425,
|
|
"epoch": 4.4911749873928395,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00035277398789534075,
|
|
"loss": 5.0294,
|
|
"mean_token_accuracy": 0.17463932186365128,
|
|
"num_tokens": 9838396.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.338879680633545,
|
|
"epoch": 4.496217851739789,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003524016523491252,
|
|
"loss": 4.976,
|
|
"mean_token_accuracy": 0.17534344792366027,
|
|
"num_tokens": 9848863.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.188317966461182,
|
|
"epoch": 4.501260716086737,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00035202907623678865,
|
|
"loss": 4.8972,
|
|
"mean_token_accuracy": 0.17871282994747162,
|
|
"num_tokens": 9860673.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.312814855575562,
|
|
"epoch": 4.506303580433686,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003516562607163062,
|
|
"loss": 4.9559,
|
|
"mean_token_accuracy": 0.1840071499347687,
|
|
"num_tokens": 9871246.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.339491271972657,
|
|
"epoch": 4.511346444780635,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003512832069463971,
|
|
"loss": 5.0191,
|
|
"mean_token_accuracy": 0.17872283309698106,
|
|
"num_tokens": 9883352.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.282098197937012,
|
|
"epoch": 4.516389309127584,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00035090991608652125,
|
|
"loss": 5.0134,
|
|
"mean_token_accuracy": 0.17903696596622468,
|
|
"num_tokens": 9894797.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.279476881027222,
|
|
"epoch": 4.521432173474533,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00035053638929687504,
|
|
"loss": 4.9442,
|
|
"mean_token_accuracy": 0.1772871270775795,
|
|
"num_tokens": 9905618.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.234744310379028,
|
|
"epoch": 4.526475037821482,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003501626277383884,
|
|
"loss": 4.9625,
|
|
"mean_token_accuracy": 0.17673694491386413,
|
|
"num_tokens": 9916817.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.269634151458741,
|
|
"epoch": 4.5315179021684315,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000349788632572721,
|
|
"loss": 4.9437,
|
|
"mean_token_accuracy": 0.1789527103304863,
|
|
"num_tokens": 9929052.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.264802598953247,
|
|
"epoch": 4.536560766515381,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003494144049622582,
|
|
"loss": 4.852,
|
|
"mean_token_accuracy": 0.17930323928594588,
|
|
"num_tokens": 9939513.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 4.536560766515381,
|
|
"eval_entropy": 5.175912131486655,
|
|
"eval_loss": 5.758546829223633,
|
|
"eval_mean_token_accuracy": 0.14862533835054353,
|
|
"eval_num_tokens": 9939513.0,
|
|
"eval_runtime": 2.6618,
|
|
"eval_samples_per_second": 1178.899,
|
|
"eval_steps_per_second": 147.644,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.179075479507446,
|
|
"epoch": 4.54160363086233,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003490399460701083,
|
|
"loss": 4.9094,
|
|
"mean_token_accuracy": 0.1804017335176468,
|
|
"num_tokens": 9950149.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.350816488265991,
|
|
"epoch": 4.546646495209279,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000348665257060098,
|
|
"loss": 4.953,
|
|
"mean_token_accuracy": 0.18287240266799926,
|
|
"num_tokens": 9961696.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.322499704360962,
|
|
"epoch": 4.551689359556228,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003482903390967695,
|
|
"loss": 4.969,
|
|
"mean_token_accuracy": 0.17883158922195436,
|
|
"num_tokens": 9971446.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.304823923110962,
|
|
"epoch": 4.556732223903177,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00034791519334537635,
|
|
"loss": 4.9211,
|
|
"mean_token_accuracy": 0.1857562929391861,
|
|
"num_tokens": 9982992.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.219585418701172,
|
|
"epoch": 4.561775088250126,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003475398209718804,
|
|
"loss": 4.8661,
|
|
"mean_token_accuracy": 0.18516680598258972,
|
|
"num_tokens": 9993443.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.2289731979370115,
|
|
"epoch": 4.566817952597075,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003471642231429475,
|
|
"loss": 4.9963,
|
|
"mean_token_accuracy": 0.1805944874882698,
|
|
"num_tokens": 10004531.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.284202766418457,
|
|
"epoch": 4.5718608169440245,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003467884010259443,
|
|
"loss": 4.9331,
|
|
"mean_token_accuracy": 0.1778637334704399,
|
|
"num_tokens": 10016455.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.291630220413208,
|
|
"epoch": 4.576903681290974,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00034641235578893476,
|
|
"loss": 4.8684,
|
|
"mean_token_accuracy": 0.19012003690004348,
|
|
"num_tokens": 10028067.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.321043825149536,
|
|
"epoch": 4.581946545637923,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003460360886006761,
|
|
"loss": 5.0684,
|
|
"mean_token_accuracy": 0.17562538534402847,
|
|
"num_tokens": 10038615.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.281301403045655,
|
|
"epoch": 4.586989409984872,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00034565960063061534,
|
|
"loss": 4.9904,
|
|
"mean_token_accuracy": 0.17908843010663986,
|
|
"num_tokens": 10050022.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.2194318771362305,
|
|
"epoch": 4.592032274331821,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003452828930488859,
|
|
"loss": 4.9261,
|
|
"mean_token_accuracy": 0.18038499653339385,
|
|
"num_tokens": 10062565.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.3913843631744385,
|
|
"epoch": 4.597075138678769,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00034490596702630355,
|
|
"loss": 5.0753,
|
|
"mean_token_accuracy": 0.17441372573375702,
|
|
"num_tokens": 10075553.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.297578525543213,
|
|
"epoch": 4.602118003025718,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003445288237343632,
|
|
"loss": 4.9867,
|
|
"mean_token_accuracy": 0.17878064662218093,
|
|
"num_tokens": 10086144.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.215610122680664,
|
|
"epoch": 4.607160867372667,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00034415146434523463,
|
|
"loss": 4.9674,
|
|
"mean_token_accuracy": 0.1801336944103241,
|
|
"num_tokens": 10097218.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.223121166229248,
|
|
"epoch": 4.6122037317196165,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00034377389003175984,
|
|
"loss": 4.9534,
|
|
"mean_token_accuracy": 0.17866341173648834,
|
|
"num_tokens": 10107614.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.380804586410522,
|
|
"epoch": 4.617246596066566,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003433961019674483,
|
|
"loss": 5.022,
|
|
"mean_token_accuracy": 0.17506256848573684,
|
|
"num_tokens": 10118976.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.3224328517913815,
|
|
"epoch": 4.622289460413515,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003430181013264741,
|
|
"loss": 4.9522,
|
|
"mean_token_accuracy": 0.17835894227027893,
|
|
"num_tokens": 10129914.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.319012546539307,
|
|
"epoch": 4.627332324760464,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003426398892836719,
|
|
"loss": 4.9607,
|
|
"mean_token_accuracy": 0.17243701219558716,
|
|
"num_tokens": 10139490.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.316942024230957,
|
|
"epoch": 4.632375189107413,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003422614670145336,
|
|
"loss": 5.0286,
|
|
"mean_token_accuracy": 0.17897747755050658,
|
|
"num_tokens": 10149874.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.287355613708496,
|
|
"epoch": 4.637418053454362,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003418828356952041,
|
|
"loss": 5.0267,
|
|
"mean_token_accuracy": 0.1783215120434761,
|
|
"num_tokens": 10161373.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.2304661750793455,
|
|
"epoch": 4.642460917801311,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00034150399650247846,
|
|
"loss": 4.9312,
|
|
"mean_token_accuracy": 0.18706550300121308,
|
|
"num_tokens": 10172044.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.295366334915161,
|
|
"epoch": 4.64750378214826,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00034112495061379764,
|
|
"loss": 4.9754,
|
|
"mean_token_accuracy": 0.17893100678920745,
|
|
"num_tokens": 10183127.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.332705879211426,
|
|
"epoch": 4.652546646495209,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00034074569920724483,
|
|
"loss": 5.0185,
|
|
"mean_token_accuracy": 0.17605760097503662,
|
|
"num_tokens": 10194366.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.298168087005616,
|
|
"epoch": 4.6575895108421586,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003403662434615423,
|
|
"loss": 4.9718,
|
|
"mean_token_accuracy": 0.1799806073307991,
|
|
"num_tokens": 10205655.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.290020704269409,
|
|
"epoch": 4.662632375189108,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00033998658455604736,
|
|
"loss": 4.9657,
|
|
"mean_token_accuracy": 0.18747677952051162,
|
|
"num_tokens": 10215596.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.238187837600708,
|
|
"epoch": 4.667675239536057,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00033960672367074827,
|
|
"loss": 4.9525,
|
|
"mean_token_accuracy": 0.17778196185827255,
|
|
"num_tokens": 10225298.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.401280212402344,
|
|
"epoch": 4.672718103883006,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003392266619862619,
|
|
"loss": 5.0672,
|
|
"mean_token_accuracy": 0.17296725660562515,
|
|
"num_tokens": 10236979.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.2960569858551025,
|
|
"epoch": 4.677760968229954,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003388464006838284,
|
|
"loss": 4.9123,
|
|
"mean_token_accuracy": 0.18287308663129806,
|
|
"num_tokens": 10248109.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.245583248138428,
|
|
"epoch": 4.682803832576903,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00033846594094530874,
|
|
"loss": 5.0003,
|
|
"mean_token_accuracy": 0.1833149090409279,
|
|
"num_tokens": 10258931.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.27817325592041,
|
|
"epoch": 4.687846696923852,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003380852839531806,
|
|
"loss": 5.0338,
|
|
"mean_token_accuracy": 0.18210602849721907,
|
|
"num_tokens": 10272521.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.318306922912598,
|
|
"epoch": 4.6928895612708015,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00033770443089053476,
|
|
"loss": 4.937,
|
|
"mean_token_accuracy": 0.1777975767850876,
|
|
"num_tokens": 10283481.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.276628589630127,
|
|
"epoch": 4.697932425617751,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00033732338294107116,
|
|
"loss": 4.992,
|
|
"mean_token_accuracy": 0.1756573885679245,
|
|
"num_tokens": 10294258.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.360388135910034,
|
|
"epoch": 4.7029752899647,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00033694214128909567,
|
|
"loss": 4.9566,
|
|
"mean_token_accuracy": 0.18480235934257508,
|
|
"num_tokens": 10304464.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.297215223312378,
|
|
"epoch": 4.708018154311649,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033656070711951633,
|
|
"loss": 4.9759,
|
|
"mean_token_accuracy": 0.17974562644958497,
|
|
"num_tokens": 10316343.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.311048603057861,
|
|
"epoch": 4.713061018658598,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003361790816178391,
|
|
"loss": 5.0346,
|
|
"mean_token_accuracy": 0.1760612115263939,
|
|
"num_tokens": 10327057.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.342336130142212,
|
|
"epoch": 4.718103883005547,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003357972659701649,
|
|
"loss": 5.0072,
|
|
"mean_token_accuracy": 0.1822867602109909,
|
|
"num_tokens": 10338483.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.261761665344238,
|
|
"epoch": 4.723146747352496,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00033541526136318564,
|
|
"loss": 4.9339,
|
|
"mean_token_accuracy": 0.18264390230178834,
|
|
"num_tokens": 10349522.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.258583879470825,
|
|
"epoch": 4.728189611699445,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003350330689841804,
|
|
"loss": 4.9123,
|
|
"mean_token_accuracy": 0.18499197959899902,
|
|
"num_tokens": 10361036.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.336564540863037,
|
|
"epoch": 4.733232476046394,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003346506900210118,
|
|
"loss": 5.0634,
|
|
"mean_token_accuracy": 0.1784227043390274,
|
|
"num_tokens": 10372673.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.306153726577759,
|
|
"epoch": 4.7382753403933435,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003342681256621227,
|
|
"loss": 5.0862,
|
|
"mean_token_accuracy": 0.17271917015314103,
|
|
"num_tokens": 10384100.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.3176604270935055,
|
|
"epoch": 4.743318204740293,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003338853770965319,
|
|
"loss": 4.9825,
|
|
"mean_token_accuracy": 0.17450401484966277,
|
|
"num_tokens": 10395503.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.3214418411254885,
|
|
"epoch": 4.748361069087242,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00033350244551383064,
|
|
"loss": 4.9399,
|
|
"mean_token_accuracy": 0.18322811126708985,
|
|
"num_tokens": 10406147.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.238266658782959,
|
|
"epoch": 4.753403933434191,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003331193321041793,
|
|
"loss": 4.8648,
|
|
"mean_token_accuracy": 0.18777066022157668,
|
|
"num_tokens": 10416659.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.286727809906006,
|
|
"epoch": 4.75844679778114,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00033273603805830296,
|
|
"loss": 4.9749,
|
|
"mean_token_accuracy": 0.17872689813375472,
|
|
"num_tokens": 10427765.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.265194940567016,
|
|
"epoch": 4.763489662128089,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00033235256456748847,
|
|
"loss": 4.9323,
|
|
"mean_token_accuracy": 0.187870055437088,
|
|
"num_tokens": 10439376.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.211277723312378,
|
|
"epoch": 4.768532526475038,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003319689128235804,
|
|
"loss": 5.0104,
|
|
"mean_token_accuracy": 0.17277956008911133,
|
|
"num_tokens": 10449232.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.193622732162476,
|
|
"epoch": 4.773575390821987,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00033158508401897707,
|
|
"loss": 4.9174,
|
|
"mean_token_accuracy": 0.18082337081432343,
|
|
"num_tokens": 10459258.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.303942489624023,
|
|
"epoch": 4.778618255168936,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003312010793466273,
|
|
"loss": 4.9856,
|
|
"mean_token_accuracy": 0.17943656146526338,
|
|
"num_tokens": 10469909.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.332628965377808,
|
|
"epoch": 4.783661119515885,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003308169000000265,
|
|
"loss": 4.9181,
|
|
"mean_token_accuracy": 0.18284550458192825,
|
|
"num_tokens": 10479848.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.309398746490478,
|
|
"epoch": 4.788703983862834,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003304325471732129,
|
|
"loss": 4.9647,
|
|
"mean_token_accuracy": 0.17774713784456253,
|
|
"num_tokens": 10490352.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.199748754501343,
|
|
"epoch": 4.793746848209783,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00033004802206076396,
|
|
"loss": 4.8914,
|
|
"mean_token_accuracy": 0.1849597916007042,
|
|
"num_tokens": 10501905.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.289130926132202,
|
|
"epoch": 4.798789712556732,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00032966332585779263,
|
|
"loss": 5.0334,
|
|
"mean_token_accuracy": 0.17518700808286666,
|
|
"num_tokens": 10512491.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.259486675262451,
|
|
"epoch": 4.803832576903681,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00032927845975994356,
|
|
"loss": 4.8401,
|
|
"mean_token_accuracy": 0.1994564414024353,
|
|
"num_tokens": 10522686.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.215784168243408,
|
|
"epoch": 4.80887544125063,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00032889342496338953,
|
|
"loss": 4.9444,
|
|
"mean_token_accuracy": 0.1773628383874893,
|
|
"num_tokens": 10534240.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.30130352973938,
|
|
"epoch": 4.813918305597579,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003285082226648273,
|
|
"loss": 5.0548,
|
|
"mean_token_accuracy": 0.17611820101737977,
|
|
"num_tokens": 10545096.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.372814464569092,
|
|
"epoch": 4.8189611699445285,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003281228540614749,
|
|
"loss": 5.032,
|
|
"mean_token_accuracy": 0.17878438234329225,
|
|
"num_tokens": 10557076.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.2200370788574215,
|
|
"epoch": 4.824004034291478,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00032773732035106657,
|
|
"loss": 4.8353,
|
|
"mean_token_accuracy": 0.186094468832016,
|
|
"num_tokens": 10567635.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.280656909942627,
|
|
"epoch": 4.829046898638427,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00032735162273185006,
|
|
"loss": 5.068,
|
|
"mean_token_accuracy": 0.16890971064567567,
|
|
"num_tokens": 10579488.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.238946580886841,
|
|
"epoch": 4.834089762985376,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003269657624025824,
|
|
"loss": 4.9279,
|
|
"mean_token_accuracy": 0.1794394999742508,
|
|
"num_tokens": 10591092.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.284884691238403,
|
|
"epoch": 4.839132627332325,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003265797405625265,
|
|
"loss": 4.9837,
|
|
"mean_token_accuracy": 0.17446444630622865,
|
|
"num_tokens": 10603165.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.202396965026855,
|
|
"epoch": 4.844175491679274,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003261935584114471,
|
|
"loss": 4.8753,
|
|
"mean_token_accuracy": 0.18996450901031495,
|
|
"num_tokens": 10615149.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.327268266677857,
|
|
"epoch": 4.849218356026223,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003258072171496072,
|
|
"loss": 5.0385,
|
|
"mean_token_accuracy": 0.17674404978752137,
|
|
"num_tokens": 10626786.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.3198000431060795,
|
|
"epoch": 4.854261220373172,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003254207179777644,
|
|
"loss": 4.9385,
|
|
"mean_token_accuracy": 0.18139078319072724,
|
|
"num_tokens": 10637107.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.255299186706543,
|
|
"epoch": 4.8593040847201205,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000325034062097167,
|
|
"loss": 4.9405,
|
|
"mean_token_accuracy": 0.18435730338096618,
|
|
"num_tokens": 10648198.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.355360269546509,
|
|
"epoch": 4.86434694906707,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00032464725070955046,
|
|
"loss": 5.0787,
|
|
"mean_token_accuracy": 0.17315272241830826,
|
|
"num_tokens": 10659867.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.336399030685425,
|
|
"epoch": 4.869389813414019,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00032426028501713347,
|
|
"loss": 4.9973,
|
|
"mean_token_accuracy": 0.17371108680963515,
|
|
"num_tokens": 10672117.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.301016807556152,
|
|
"epoch": 4.874432677760968,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00032387316622261425,
|
|
"loss": 4.8853,
|
|
"mean_token_accuracy": 0.18303302377462388,
|
|
"num_tokens": 10682014.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.277377986907959,
|
|
"epoch": 4.879475542107917,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00032348589552916703,
|
|
"loss": 5.0683,
|
|
"mean_token_accuracy": 0.16809291690587996,
|
|
"num_tokens": 10693330.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.2971728324890135,
|
|
"epoch": 4.884518406454866,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003230984741404381,
|
|
"loss": 4.9142,
|
|
"mean_token_accuracy": 0.1854029804468155,
|
|
"num_tokens": 10703772.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.313681793212891,
|
|
"epoch": 4.889561270801815,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003227109032605419,
|
|
"loss": 5.0097,
|
|
"mean_token_accuracy": 0.17449059784412385,
|
|
"num_tokens": 10714088.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.184208488464355,
|
|
"epoch": 4.894604135148764,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003223231840940578,
|
|
"loss": 4.8,
|
|
"mean_token_accuracy": 0.19187837094068527,
|
|
"num_tokens": 10724701.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.263138484954834,
|
|
"epoch": 4.8996469994957135,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00032193531784602595,
|
|
"loss": 4.9382,
|
|
"mean_token_accuracy": 0.18246944397687911,
|
|
"num_tokens": 10735361.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.332172155380249,
|
|
"epoch": 4.904689863842663,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00032154730572194346,
|
|
"loss": 4.9422,
|
|
"mean_token_accuracy": 0.17696973085403442,
|
|
"num_tokens": 10745909.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.26082067489624,
|
|
"epoch": 4.909732728189612,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000321159148927761,
|
|
"loss": 5.0684,
|
|
"mean_token_accuracy": 0.172713965177536,
|
|
"num_tokens": 10759305.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.268248128890991,
|
|
"epoch": 4.914775592536561,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000320770848669879,
|
|
"loss": 5.0205,
|
|
"mean_token_accuracy": 0.18507444709539414,
|
|
"num_tokens": 10770451.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.285018444061279,
|
|
"epoch": 4.91981845688351,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00032038240615514345,
|
|
"loss": 4.9394,
|
|
"mean_token_accuracy": 0.18130892664194107,
|
|
"num_tokens": 10782376.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.27263913154602,
|
|
"epoch": 4.924861321230459,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003199938225908426,
|
|
"loss": 5.0551,
|
|
"mean_token_accuracy": 0.1728331208229065,
|
|
"num_tokens": 10794252.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.31664514541626,
|
|
"epoch": 4.929904185577408,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003196050991847033,
|
|
"loss": 4.9955,
|
|
"mean_token_accuracy": 0.1778052940964699,
|
|
"num_tokens": 10804117.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.289377689361572,
|
|
"epoch": 4.934947049924357,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003192162371448868,
|
|
"loss": 4.93,
|
|
"mean_token_accuracy": 0.17900618761777878,
|
|
"num_tokens": 10815798.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.240103912353516,
|
|
"epoch": 4.939989914271306,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00031882723767998517,
|
|
"loss": 4.9498,
|
|
"mean_token_accuracy": 0.18601714968681335,
|
|
"num_tokens": 10827135.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.301865100860596,
|
|
"epoch": 4.9450327786182555,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00031843810199901795,
|
|
"loss": 5.0496,
|
|
"mean_token_accuracy": 0.17293065041303635,
|
|
"num_tokens": 10838988.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.265445232391357,
|
|
"epoch": 4.950075642965205,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003180488313114277,
|
|
"loss": 4.8719,
|
|
"mean_token_accuracy": 0.18641145378351212,
|
|
"num_tokens": 10848223.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.273135089874268,
|
|
"epoch": 4.955118507312154,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003176594268270766,
|
|
"loss": 4.9127,
|
|
"mean_token_accuracy": 0.1829720050096512,
|
|
"num_tokens": 10859052.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.261722087860107,
|
|
"epoch": 4.960161371659103,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00031726988975624293,
|
|
"loss": 4.9687,
|
|
"mean_token_accuracy": 0.17940526008605956,
|
|
"num_tokens": 10869108.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.276017951965332,
|
|
"epoch": 4.965204236006051,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00031688022130961675,
|
|
"loss": 4.9746,
|
|
"mean_token_accuracy": 0.1809917390346527,
|
|
"num_tokens": 10880691.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.227404642105102,
|
|
"epoch": 4.970247100353,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003164904226982967,
|
|
"loss": 4.8961,
|
|
"mean_token_accuracy": 0.1789270132780075,
|
|
"num_tokens": 10892206.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.25455961227417,
|
|
"epoch": 4.975289964699949,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00031610049513378574,
|
|
"loss": 4.8546,
|
|
"mean_token_accuracy": 0.19265048652887345,
|
|
"num_tokens": 10902130.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.30515718460083,
|
|
"epoch": 4.9803328290468984,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00031571043982798786,
|
|
"loss": 4.9953,
|
|
"mean_token_accuracy": 0.17955620884895324,
|
|
"num_tokens": 10913424.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.267713642120361,
|
|
"epoch": 4.985375693393848,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003153202579932039,
|
|
"loss": 5.0561,
|
|
"mean_token_accuracy": 0.17732520550489425,
|
|
"num_tokens": 10925718.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.286071157455444,
|
|
"epoch": 4.990418557740797,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00031492995084212794,
|
|
"loss": 4.9279,
|
|
"mean_token_accuracy": 0.1820734038949013,
|
|
"num_tokens": 10937102.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.263105726242065,
|
|
"epoch": 4.995461422087746,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003145395195878437,
|
|
"loss": 4.926,
|
|
"mean_token_accuracy": 0.18130062371492386,
|
|
"num_tokens": 10948679.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.197415828704834,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.00031414896544382047,
|
|
"loss": 4.856,
|
|
"mean_token_accuracy": 0.18261487616433036,
|
|
"num_tokens": 10958200.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.2683610916137695,
|
|
"epoch": 5.005042864346949,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003137582896239095,
|
|
"loss": 4.6695,
|
|
"mean_token_accuracy": 0.2016356647014618,
|
|
"num_tokens": 10969775.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.221837902069092,
|
|
"epoch": 5.010085728693898,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003133674933423404,
|
|
"loss": 4.653,
|
|
"mean_token_accuracy": 0.19351939409971236,
|
|
"num_tokens": 10980423.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.192023706436157,
|
|
"epoch": 5.015128593040847,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00031297657781371687,
|
|
"loss": 4.6842,
|
|
"mean_token_accuracy": 0.19781454056501388,
|
|
"num_tokens": 10992716.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.2818540096282955,
|
|
"epoch": 5.0201714573877965,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003125855442530135,
|
|
"loss": 4.7581,
|
|
"mean_token_accuracy": 0.19106778204441072,
|
|
"num_tokens": 11004597.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.31525297164917,
|
|
"epoch": 5.025214321734746,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00031219439387557155,
|
|
"loss": 4.7219,
|
|
"mean_token_accuracy": 0.19655250757932663,
|
|
"num_tokens": 11015423.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.26198263168335,
|
|
"epoch": 5.030257186081695,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00031180312789709535,
|
|
"loss": 4.7544,
|
|
"mean_token_accuracy": 0.2012348875403404,
|
|
"num_tokens": 11028191.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.1872289180755615,
|
|
"epoch": 5.035300050428644,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003114117475336487,
|
|
"loss": 4.772,
|
|
"mean_token_accuracy": 0.1994599238038063,
|
|
"num_tokens": 11040690.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.230694532394409,
|
|
"epoch": 5.040342914775593,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003110202540016509,
|
|
"loss": 4.6281,
|
|
"mean_token_accuracy": 0.20158670991659164,
|
|
"num_tokens": 11052580.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 5.040342914775593,
|
|
"eval_entropy": 5.203414428021769,
|
|
"eval_loss": 5.734892845153809,
|
|
"eval_mean_token_accuracy": 0.1512179922208228,
|
|
"eval_num_tokens": 11052580.0,
|
|
"eval_runtime": 2.6477,
|
|
"eval_samples_per_second": 1185.178,
|
|
"eval_steps_per_second": 148.431,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.244221115112305,
|
|
"epoch": 5.045385779122542,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00031062864851787267,
|
|
"loss": 4.6772,
|
|
"mean_token_accuracy": 0.19672164916992188,
|
|
"num_tokens": 11063249.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.215124845504761,
|
|
"epoch": 5.05042864346949,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00031023693229943294,
|
|
"loss": 4.6594,
|
|
"mean_token_accuracy": 0.1915979117155075,
|
|
"num_tokens": 11073738.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.179732704162598,
|
|
"epoch": 5.055471507816439,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003098451065637948,
|
|
"loss": 4.6993,
|
|
"mean_token_accuracy": 0.1922016367316246,
|
|
"num_tokens": 11084555.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.254594564437866,
|
|
"epoch": 5.0605143721633885,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00030945317252876163,
|
|
"loss": 4.6914,
|
|
"mean_token_accuracy": 0.20336001366376877,
|
|
"num_tokens": 11095947.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.227708196640014,
|
|
"epoch": 5.065557236510338,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00030906113141247353,
|
|
"loss": 4.6424,
|
|
"mean_token_accuracy": 0.20187090933322907,
|
|
"num_tokens": 11106332.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.230111789703369,
|
|
"epoch": 5.070600100857287,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00030866898443340324,
|
|
"loss": 4.7601,
|
|
"mean_token_accuracy": 0.2000565320253372,
|
|
"num_tokens": 11118396.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.253856086730957,
|
|
"epoch": 5.075642965204236,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003082767328103527,
|
|
"loss": 4.6721,
|
|
"mean_token_accuracy": 0.19586574286222458,
|
|
"num_tokens": 11128702.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.201744604110718,
|
|
"epoch": 5.080685829551185,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00030788437776244895,
|
|
"loss": 4.6669,
|
|
"mean_token_accuracy": 0.19566670805215836,
|
|
"num_tokens": 11138004.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.186066484451294,
|
|
"epoch": 5.085728693898134,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003074919205091405,
|
|
"loss": 4.6978,
|
|
"mean_token_accuracy": 0.2027225002646446,
|
|
"num_tokens": 11149320.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.27611780166626,
|
|
"epoch": 5.090771558245083,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003070993622701936,
|
|
"loss": 4.7149,
|
|
"mean_token_accuracy": 0.1926956444978714,
|
|
"num_tokens": 11160942.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.223322248458862,
|
|
"epoch": 5.095814422592032,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00030670670426568837,
|
|
"loss": 4.6757,
|
|
"mean_token_accuracy": 0.1979008585214615,
|
|
"num_tokens": 11172109.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.232969141006469,
|
|
"epoch": 5.100857286938981,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003063139477160147,
|
|
"loss": 4.7118,
|
|
"mean_token_accuracy": 0.19624906927347183,
|
|
"num_tokens": 11182455.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.204172945022583,
|
|
"epoch": 5.1059001512859306,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00030592109384186935,
|
|
"loss": 4.7842,
|
|
"mean_token_accuracy": 0.19230634570121766,
|
|
"num_tokens": 11194612.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.244259166717529,
|
|
"epoch": 5.11094301563288,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000305528143864251,
|
|
"loss": 4.683,
|
|
"mean_token_accuracy": 0.20314915627241134,
|
|
"num_tokens": 11205889.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.193525505065918,
|
|
"epoch": 5.115985879979829,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00030513509900445725,
|
|
"loss": 4.6372,
|
|
"mean_token_accuracy": 0.20497628897428513,
|
|
"num_tokens": 11217386.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.226591634750366,
|
|
"epoch": 5.121028744326778,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00030474196048408065,
|
|
"loss": 4.7582,
|
|
"mean_token_accuracy": 0.19505153149366378,
|
|
"num_tokens": 11229963.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.205451774597168,
|
|
"epoch": 5.126071608673727,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003043487295250049,
|
|
"loss": 4.6976,
|
|
"mean_token_accuracy": 0.2017916962504387,
|
|
"num_tokens": 11240959.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.140781736373901,
|
|
"epoch": 5.131114473020676,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003039554073494006,
|
|
"loss": 4.6032,
|
|
"mean_token_accuracy": 0.2034008637070656,
|
|
"num_tokens": 11251971.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.218955135345459,
|
|
"epoch": 5.136157337367624,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003035619951797224,
|
|
"loss": 4.7152,
|
|
"mean_token_accuracy": 0.1943013146519661,
|
|
"num_tokens": 11261933.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.249277877807617,
|
|
"epoch": 5.1412002017145735,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003031684942387043,
|
|
"loss": 4.783,
|
|
"mean_token_accuracy": 0.1915792256593704,
|
|
"num_tokens": 11272579.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.282430791854859,
|
|
"epoch": 5.146243066061523,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00030277490574935633,
|
|
"loss": 4.7694,
|
|
"mean_token_accuracy": 0.19303622990846633,
|
|
"num_tokens": 11282675.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.254349851608277,
|
|
"epoch": 5.151285930408472,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003023812309349605,
|
|
"loss": 4.7523,
|
|
"mean_token_accuracy": 0.18679001480340957,
|
|
"num_tokens": 11294837.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.306874322891235,
|
|
"epoch": 5.156328794755421,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003019874710190674,
|
|
"loss": 4.818,
|
|
"mean_token_accuracy": 0.18909795731306075,
|
|
"num_tokens": 11305723.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.302325630187989,
|
|
"epoch": 5.16137165910237,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003015936272254918,
|
|
"loss": 4.7727,
|
|
"mean_token_accuracy": 0.19244884252548217,
|
|
"num_tokens": 11316034.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.174959802627564,
|
|
"epoch": 5.166414523449319,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003011997007783094,
|
|
"loss": 4.6763,
|
|
"mean_token_accuracy": 0.19987003058195113,
|
|
"num_tokens": 11325515.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.2731081485748295,
|
|
"epoch": 5.171457387796268,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003008056929018525,
|
|
"loss": 4.7402,
|
|
"mean_token_accuracy": 0.19476868212223053,
|
|
"num_tokens": 11336874.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.223803472518921,
|
|
"epoch": 5.176500252143217,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00030041160482070686,
|
|
"loss": 4.7615,
|
|
"mean_token_accuracy": 0.19219011962413787,
|
|
"num_tokens": 11347749.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.1939005851745605,
|
|
"epoch": 5.181543116490166,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00030001743775970714,
|
|
"loss": 4.7494,
|
|
"mean_token_accuracy": 0.1913871258497238,
|
|
"num_tokens": 11358865.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.173363208770752,
|
|
"epoch": 5.1865859808371155,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00029962319294393383,
|
|
"loss": 4.7165,
|
|
"mean_token_accuracy": 0.19477847814559937,
|
|
"num_tokens": 11369699.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.198025846481324,
|
|
"epoch": 5.191628845184065,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002992288715987088,
|
|
"loss": 4.705,
|
|
"mean_token_accuracy": 0.2007056951522827,
|
|
"num_tokens": 11381532.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.268914699554443,
|
|
"epoch": 5.196671709531014,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002988344749495919,
|
|
"loss": 4.7319,
|
|
"mean_token_accuracy": 0.19766185134649278,
|
|
"num_tokens": 11392373.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.248008441925049,
|
|
"epoch": 5.201714573877963,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00029844000422237714,
|
|
"loss": 4.7541,
|
|
"mean_token_accuracy": 0.19771358072757722,
|
|
"num_tokens": 11403732.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.203286075592041,
|
|
"epoch": 5.206757438224912,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00029804546064308854,
|
|
"loss": 4.6873,
|
|
"mean_token_accuracy": 0.20012180358171464,
|
|
"num_tokens": 11415344.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.163456201553345,
|
|
"epoch": 5.211800302571861,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00029765084543797657,
|
|
"loss": 4.7029,
|
|
"mean_token_accuracy": 0.19341932684183122,
|
|
"num_tokens": 11427564.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.246589231491089,
|
|
"epoch": 5.21684316691881,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002972561598335145,
|
|
"loss": 4.7402,
|
|
"mean_token_accuracy": 0.1988281786441803,
|
|
"num_tokens": 11439119.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.222621583938599,
|
|
"epoch": 5.221886031265759,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002968614050563943,
|
|
"loss": 4.7605,
|
|
"mean_token_accuracy": 0.19533331245183944,
|
|
"num_tokens": 11450460.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.227239179611206,
|
|
"epoch": 5.2269288956127085,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002964665823335229,
|
|
"loss": 4.7793,
|
|
"mean_token_accuracy": 0.1891628012061119,
|
|
"num_tokens": 11461414.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.227170467376709,
|
|
"epoch": 5.231971759959657,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00029607169289201836,
|
|
"loss": 4.7577,
|
|
"mean_token_accuracy": 0.1979129508137703,
|
|
"num_tokens": 11472071.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.275568199157715,
|
|
"epoch": 5.237014624306606,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00029567673795920625,
|
|
"loss": 4.7526,
|
|
"mean_token_accuracy": 0.19085296541452407,
|
|
"num_tokens": 11483098.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.221810054779053,
|
|
"epoch": 5.242057488653555,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00029528171876261555,
|
|
"loss": 4.7211,
|
|
"mean_token_accuracy": 0.1987485721707344,
|
|
"num_tokens": 11494339.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.241558933258057,
|
|
"epoch": 5.247100353000504,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002948866365299751,
|
|
"loss": 4.8055,
|
|
"mean_token_accuracy": 0.1881676584482193,
|
|
"num_tokens": 11504539.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.261566781997681,
|
|
"epoch": 5.252143217347453,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002944914924892096,
|
|
"loss": 4.8025,
|
|
"mean_token_accuracy": 0.18994247764348984,
|
|
"num_tokens": 11515298.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.200396347045898,
|
|
"epoch": 5.257186081694402,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002940962878684358,
|
|
"loss": 4.6099,
|
|
"mean_token_accuracy": 0.20612432360649108,
|
|
"num_tokens": 11524766.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.208489418029785,
|
|
"epoch": 5.262228946041351,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002937010238959588,
|
|
"loss": 4.7801,
|
|
"mean_token_accuracy": 0.18987920135259628,
|
|
"num_tokens": 11536603.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.201207685470581,
|
|
"epoch": 5.2672718103883005,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002933057018002681,
|
|
"loss": 4.8119,
|
|
"mean_token_accuracy": 0.19679956138134003,
|
|
"num_tokens": 11548535.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.229961109161377,
|
|
"epoch": 5.27231467473525,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000292910322810034,
|
|
"loss": 4.7097,
|
|
"mean_token_accuracy": 0.19394957423210143,
|
|
"num_tokens": 11559044.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.167344284057617,
|
|
"epoch": 5.277357539082199,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00029251488815410344,
|
|
"loss": 4.7195,
|
|
"mean_token_accuracy": 0.19605668038129806,
|
|
"num_tokens": 11571276.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.193845748901367,
|
|
"epoch": 5.282400403429148,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00029211939906149644,
|
|
"loss": 4.7258,
|
|
"mean_token_accuracy": 0.1945561096072197,
|
|
"num_tokens": 11581935.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.22123703956604,
|
|
"epoch": 5.287443267776097,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00029172385676140225,
|
|
"loss": 4.8348,
|
|
"mean_token_accuracy": 0.19095736891031265,
|
|
"num_tokens": 11594188.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.260570383071899,
|
|
"epoch": 5.292486132123046,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00029132826248317537,
|
|
"loss": 4.8277,
|
|
"mean_token_accuracy": 0.18733905851840973,
|
|
"num_tokens": 11606025.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.2465917587280275,
|
|
"epoch": 5.297528996469995,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002909326174563322,
|
|
"loss": 4.7588,
|
|
"mean_token_accuracy": 0.19410619139671326,
|
|
"num_tokens": 11617660.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.2093010425567625,
|
|
"epoch": 5.302571860816944,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00029053692291054636,
|
|
"loss": 4.7329,
|
|
"mean_token_accuracy": 0.19946946948766708,
|
|
"num_tokens": 11628137.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.198204421997071,
|
|
"epoch": 5.307614725163893,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00029014118007564564,
|
|
"loss": 4.7862,
|
|
"mean_token_accuracy": 0.18725667297840118,
|
|
"num_tokens": 11639123.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.237512302398682,
|
|
"epoch": 5.3126575895108425,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002897453901816081,
|
|
"loss": 4.7571,
|
|
"mean_token_accuracy": 0.2011867120862007,
|
|
"num_tokens": 11650998.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.204316282272339,
|
|
"epoch": 5.317700453857791,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00028934955445855773,
|
|
"loss": 4.726,
|
|
"mean_token_accuracy": 0.19807681441307068,
|
|
"num_tokens": 11660543.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.2303680896759035,
|
|
"epoch": 5.32274331820474,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00028895367413676095,
|
|
"loss": 4.7263,
|
|
"mean_token_accuracy": 0.19477445334196092,
|
|
"num_tokens": 11671258.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.213824701309204,
|
|
"epoch": 5.327786182551689,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002885577504466232,
|
|
"loss": 4.7607,
|
|
"mean_token_accuracy": 0.1981363207101822,
|
|
"num_tokens": 11682498.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.218505859375,
|
|
"epoch": 5.332829046898638,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00028816178461868425,
|
|
"loss": 4.7676,
|
|
"mean_token_accuracy": 0.19133229106664656,
|
|
"num_tokens": 11694818.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.215164279937744,
|
|
"epoch": 5.337871911245587,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002877657778836151,
|
|
"loss": 4.7646,
|
|
"mean_token_accuracy": 0.18880323469638824,
|
|
"num_tokens": 11705939.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.232865810394287,
|
|
"epoch": 5.342914775592536,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002873697314722138,
|
|
"loss": 4.793,
|
|
"mean_token_accuracy": 0.19156908690929414,
|
|
"num_tokens": 11716885.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.261131858825683,
|
|
"epoch": 5.3479576399394855,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002869736466154018,
|
|
"loss": 4.7777,
|
|
"mean_token_accuracy": 0.1926743283867836,
|
|
"num_tokens": 11728423.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.203846788406372,
|
|
"epoch": 5.353000504286435,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00028657752454421974,
|
|
"loss": 4.7611,
|
|
"mean_token_accuracy": 0.19454611986875534,
|
|
"num_tokens": 11739293.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.274653530120849,
|
|
"epoch": 5.358043368633384,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002861813664898245,
|
|
"loss": 4.7543,
|
|
"mean_token_accuracy": 0.1915184125304222,
|
|
"num_tokens": 11749915.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.1688720703125,
|
|
"epoch": 5.363086232980333,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00028578517368348425,
|
|
"loss": 4.7928,
|
|
"mean_token_accuracy": 0.19510262459516525,
|
|
"num_tokens": 11760055.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.157090282440185,
|
|
"epoch": 5.368129097327282,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002853889473565753,
|
|
"loss": 4.7649,
|
|
"mean_token_accuracy": 0.19502191096544266,
|
|
"num_tokens": 11771354.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.237582683563232,
|
|
"epoch": 5.373171961674231,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002849926887405784,
|
|
"loss": 4.8004,
|
|
"mean_token_accuracy": 0.19729410111904144,
|
|
"num_tokens": 11782628.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.182104587554932,
|
|
"epoch": 5.37821482602118,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002845963990670744,
|
|
"loss": 4.7201,
|
|
"mean_token_accuracy": 0.19249078631401062,
|
|
"num_tokens": 11794733.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.316960573196411,
|
|
"epoch": 5.383257690368129,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00028420007956774077,
|
|
"loss": 4.831,
|
|
"mean_token_accuracy": 0.18851348459720613,
|
|
"num_tokens": 11806532.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.21695466041565,
|
|
"epoch": 5.388300554715078,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00028380373147434746,
|
|
"loss": 4.7737,
|
|
"mean_token_accuracy": 0.18902404010295867,
|
|
"num_tokens": 11817305.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.177695226669312,
|
|
"epoch": 5.3933434190620275,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00028340735601875376,
|
|
"loss": 4.6833,
|
|
"mean_token_accuracy": 0.20029393434524537,
|
|
"num_tokens": 11827400.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.195882844924927,
|
|
"epoch": 5.398386283408977,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002830109544329036,
|
|
"loss": 4.7534,
|
|
"mean_token_accuracy": 0.19227079600095748,
|
|
"num_tokens": 11837675.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.113247060775757,
|
|
"epoch": 5.403429147755926,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00028261452794882223,
|
|
"loss": 4.6331,
|
|
"mean_token_accuracy": 0.2053363084793091,
|
|
"num_tokens": 11846567.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.272407579421997,
|
|
"epoch": 5.408472012102875,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00028221807779861243,
|
|
"loss": 4.7983,
|
|
"mean_token_accuracy": 0.1971825897693634,
|
|
"num_tokens": 11856945.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.2371735095977785,
|
|
"epoch": 5.413514876449823,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002818216052144502,
|
|
"loss": 4.7159,
|
|
"mean_token_accuracy": 0.2004765510559082,
|
|
"num_tokens": 11867302.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.214233636856079,
|
|
"epoch": 5.418557740796772,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002814251114285816,
|
|
"loss": 4.8016,
|
|
"mean_token_accuracy": 0.191142500936985,
|
|
"num_tokens": 11879351.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.159127330780029,
|
|
"epoch": 5.423600605143721,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002810285976733186,
|
|
"loss": 4.7367,
|
|
"mean_token_accuracy": 0.19349124133586884,
|
|
"num_tokens": 11890337.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.224794864654541,
|
|
"epoch": 5.4286434694906704,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002806320651810349,
|
|
"loss": 4.7268,
|
|
"mean_token_accuracy": 0.1979726195335388,
|
|
"num_tokens": 11900931.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.280755186080933,
|
|
"epoch": 5.43368633383762,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002802355151841627,
|
|
"loss": 4.7631,
|
|
"mean_token_accuracy": 0.1947193518280983,
|
|
"num_tokens": 11911901.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.130286312103271,
|
|
"epoch": 5.438729198184569,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002798389489151886,
|
|
"loss": 4.7691,
|
|
"mean_token_accuracy": 0.20049044191837312,
|
|
"num_tokens": 11923349.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.149282217025757,
|
|
"epoch": 5.443772062531518,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00027944236760664947,
|
|
"loss": 4.6871,
|
|
"mean_token_accuracy": 0.19646586179733277,
|
|
"num_tokens": 11933496.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.265764284133911,
|
|
"epoch": 5.448814926878467,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00027904577249112945,
|
|
"loss": 4.8004,
|
|
"mean_token_accuracy": 0.19084361344575881,
|
|
"num_tokens": 11944626.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.241815662384033,
|
|
"epoch": 5.453857791225416,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002786491648012552,
|
|
"loss": 4.7374,
|
|
"mean_token_accuracy": 0.19580766409635544,
|
|
"num_tokens": 11956588.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.2382903575897215,
|
|
"epoch": 5.458900655572365,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002782525457696926,
|
|
"loss": 4.7706,
|
|
"mean_token_accuracy": 0.19070635735988617,
|
|
"num_tokens": 11967375.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.178342723846436,
|
|
"epoch": 5.463943519919314,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002778559166291427,
|
|
"loss": 4.723,
|
|
"mean_token_accuracy": 0.19274348467588426,
|
|
"num_tokens": 11980651.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.189646053314209,
|
|
"epoch": 5.468986384266263,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002774592786123379,
|
|
"loss": 4.7035,
|
|
"mean_token_accuracy": 0.1992602601647377,
|
|
"num_tokens": 11991727.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.198212385177612,
|
|
"epoch": 5.4740292486132125,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002770626329520386,
|
|
"loss": 4.7231,
|
|
"mean_token_accuracy": 0.19862120002508163,
|
|
"num_tokens": 12001820.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.236854553222656,
|
|
"epoch": 5.479072112960162,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002766659808810285,
|
|
"loss": 4.7371,
|
|
"mean_token_accuracy": 0.19178324490785598,
|
|
"num_tokens": 12012196.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.218763971328736,
|
|
"epoch": 5.484114977307111,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00027626932363211134,
|
|
"loss": 4.7654,
|
|
"mean_token_accuracy": 0.18559939414262772,
|
|
"num_tokens": 12022701.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.191195297241211,
|
|
"epoch": 5.48915784165406,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002758726624381071,
|
|
"loss": 4.7249,
|
|
"mean_token_accuracy": 0.20049473345279695,
|
|
"num_tokens": 12032773.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.221637201309204,
|
|
"epoch": 5.494200706001009,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002754759985318478,
|
|
"loss": 4.7804,
|
|
"mean_token_accuracy": 0.19656205028295518,
|
|
"num_tokens": 12044140.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.254353284835815,
|
|
"epoch": 5.499243570347957,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002750793331461741,
|
|
"loss": 4.7973,
|
|
"mean_token_accuracy": 0.18787062913179398,
|
|
"num_tokens": 12055680.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.2182573795318605,
|
|
"epoch": 5.504286434694906,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002746826675139313,
|
|
"loss": 4.7533,
|
|
"mean_token_accuracy": 0.1958657756447792,
|
|
"num_tokens": 12068107.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.1608528137207035,
|
|
"epoch": 5.509329299041855,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00027428600286796503,
|
|
"loss": 4.7121,
|
|
"mean_token_accuracy": 0.1978359118103981,
|
|
"num_tokens": 12079242.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.2175439357757565,
|
|
"epoch": 5.5143721633888045,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00027388934044111836,
|
|
"loss": 4.7748,
|
|
"mean_token_accuracy": 0.1902337282896042,
|
|
"num_tokens": 12090720.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.192400789260864,
|
|
"epoch": 5.519415027735754,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002734926814662271,
|
|
"loss": 4.7125,
|
|
"mean_token_accuracy": 0.19794080257415772,
|
|
"num_tokens": 12101168.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.194960737228394,
|
|
"epoch": 5.524457892082703,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00027309602717611673,
|
|
"loss": 4.7869,
|
|
"mean_token_accuracy": 0.19182594120502472,
|
|
"num_tokens": 12112233.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.238377571105957,
|
|
"epoch": 5.529500756429652,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00027269937880359786,
|
|
"loss": 4.7781,
|
|
"mean_token_accuracy": 0.18964944928884506,
|
|
"num_tokens": 12122243.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.294060087203979,
|
|
"epoch": 5.534543620776601,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00027230273758146263,
|
|
"loss": 4.7997,
|
|
"mean_token_accuracy": 0.18920252323150635,
|
|
"num_tokens": 12132164.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.2104791641235355,
|
|
"epoch": 5.53958648512355,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002719061047424813,
|
|
"loss": 4.6719,
|
|
"mean_token_accuracy": 0.20316760390996932,
|
|
"num_tokens": 12141490.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.177541208267212,
|
|
"epoch": 5.544629349470499,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002715094815193978,
|
|
"loss": 4.7431,
|
|
"mean_token_accuracy": 0.18889395147562027,
|
|
"num_tokens": 12151772.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 5.544629349470499,
|
|
"eval_entropy": 5.105624135820617,
|
|
"eval_loss": 5.72531270980835,
|
|
"eval_mean_token_accuracy": 0.15254807381229546,
|
|
"eval_num_tokens": 12151772.0,
|
|
"eval_runtime": 2.6618,
|
|
"eval_samples_per_second": 1178.912,
|
|
"eval_steps_per_second": 147.646,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.155550909042359,
|
|
"epoch": 5.549672213817448,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002711128691449262,
|
|
"loss": 4.785,
|
|
"mean_token_accuracy": 0.1964417055249214,
|
|
"num_tokens": 12163398.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.229569625854492,
|
|
"epoch": 5.5547150781643975,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000270716268851747,
|
|
"loss": 4.8276,
|
|
"mean_token_accuracy": 0.1893754005432129,
|
|
"num_tokens": 12173809.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.253583526611328,
|
|
"epoch": 5.559757942511347,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00027031968187250305,
|
|
"loss": 4.8092,
|
|
"mean_token_accuracy": 0.18427780270576477,
|
|
"num_tokens": 12185366.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.278695774078369,
|
|
"epoch": 5.564800806858296,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002699231094397957,
|
|
"loss": 4.7108,
|
|
"mean_token_accuracy": 0.1966024249792099,
|
|
"num_tokens": 12197613.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.191903877258301,
|
|
"epoch": 5.569843671205245,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002695265527861815,
|
|
"loss": 4.7664,
|
|
"mean_token_accuracy": 0.1918341413140297,
|
|
"num_tokens": 12209152.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.166343879699707,
|
|
"epoch": 5.574886535552194,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002691300131441674,
|
|
"loss": 4.6656,
|
|
"mean_token_accuracy": 0.20030797123908997,
|
|
"num_tokens": 12219879.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.181400299072266,
|
|
"epoch": 5.579929399899143,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002687334917462079,
|
|
"loss": 4.7912,
|
|
"mean_token_accuracy": 0.1872957706451416,
|
|
"num_tokens": 12231254.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.145044279098511,
|
|
"epoch": 5.584972264246092,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002683369898247005,
|
|
"loss": 4.6892,
|
|
"mean_token_accuracy": 0.19938318729400634,
|
|
"num_tokens": 12241545.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.121308708190918,
|
|
"epoch": 5.590015128593041,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002679405086119825,
|
|
"loss": 4.6855,
|
|
"mean_token_accuracy": 0.1969492882490158,
|
|
"num_tokens": 12252642.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.2301459312438965,
|
|
"epoch": 5.59505799293999,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00026754404934032657,
|
|
"loss": 4.7923,
|
|
"mean_token_accuracy": 0.19351887851953506,
|
|
"num_tokens": 12264296.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.2317241668701175,
|
|
"epoch": 5.600100857286939,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002671476132419374,
|
|
"loss": 4.731,
|
|
"mean_token_accuracy": 0.19715087860822678,
|
|
"num_tokens": 12274127.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.209739780426025,
|
|
"epoch": 5.605143721633888,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002667512015489472,
|
|
"loss": 4.7738,
|
|
"mean_token_accuracy": 0.1959247276186943,
|
|
"num_tokens": 12287333.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.225480556488037,
|
|
"epoch": 5.610186585980837,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000266354815493413,
|
|
"loss": 4.711,
|
|
"mean_token_accuracy": 0.2012144699692726,
|
|
"num_tokens": 12297172.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.204460000991821,
|
|
"epoch": 5.615229450327786,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00026595845630731166,
|
|
"loss": 4.7553,
|
|
"mean_token_accuracy": 0.1954453781247139,
|
|
"num_tokens": 12308505.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.140445756912231,
|
|
"epoch": 5.620272314674735,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00026556212522253666,
|
|
"loss": 4.6918,
|
|
"mean_token_accuracy": 0.19891340732574464,
|
|
"num_tokens": 12319108.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.184196996688843,
|
|
"epoch": 5.625315179021684,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002651658234708941,
|
|
"loss": 4.7646,
|
|
"mean_token_accuracy": 0.19798631519079207,
|
|
"num_tokens": 12331935.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.21385178565979,
|
|
"epoch": 5.630358043368633,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00026476955228409924,
|
|
"loss": 4.7394,
|
|
"mean_token_accuracy": 0.19849587827920914,
|
|
"num_tokens": 12343380.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.212689971923828,
|
|
"epoch": 5.635400907715582,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00026437331289377175,
|
|
"loss": 4.7692,
|
|
"mean_token_accuracy": 0.19517802447080612,
|
|
"num_tokens": 12355216.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.255007410049439,
|
|
"epoch": 5.640443772062532,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002639771065314329,
|
|
"loss": 4.838,
|
|
"mean_token_accuracy": 0.18861068934202194,
|
|
"num_tokens": 12367339.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.1903173446655275,
|
|
"epoch": 5.645486636409481,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00026358093442850134,
|
|
"loss": 4.775,
|
|
"mean_token_accuracy": 0.1912614032626152,
|
|
"num_tokens": 12378427.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.199225187301636,
|
|
"epoch": 5.65052950075643,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002631847978162889,
|
|
"loss": 4.722,
|
|
"mean_token_accuracy": 0.19337614476680756,
|
|
"num_tokens": 12389264.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.1688987731933596,
|
|
"epoch": 5.655572365103379,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00026278869792599745,
|
|
"loss": 4.7833,
|
|
"mean_token_accuracy": 0.19639360159635544,
|
|
"num_tokens": 12400080.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.22928409576416,
|
|
"epoch": 5.660615229450328,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002623926359887145,
|
|
"loss": 4.7674,
|
|
"mean_token_accuracy": 0.1899639382958412,
|
|
"num_tokens": 12410996.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.212386560440064,
|
|
"epoch": 5.665658093797277,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00026199661323540973,
|
|
"loss": 4.7706,
|
|
"mean_token_accuracy": 0.19670359045267105,
|
|
"num_tokens": 12421960.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.204670810699463,
|
|
"epoch": 5.670700958144226,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00026160063089693096,
|
|
"loss": 4.809,
|
|
"mean_token_accuracy": 0.19123040288686752,
|
|
"num_tokens": 12433254.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.181452798843384,
|
|
"epoch": 5.675743822491175,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002612046902040006,
|
|
"loss": 4.6429,
|
|
"mean_token_accuracy": 0.20305237472057341,
|
|
"num_tokens": 12443278.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.194872426986694,
|
|
"epoch": 5.680786686838124,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002608087923872111,
|
|
"loss": 4.7799,
|
|
"mean_token_accuracy": 0.19284093827009202,
|
|
"num_tokens": 12456228.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.152115774154663,
|
|
"epoch": 5.685829551185073,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00026041293867702235,
|
|
"loss": 4.7162,
|
|
"mean_token_accuracy": 0.19667498767375946,
|
|
"num_tokens": 12467041.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.189820098876953,
|
|
"epoch": 5.690872415532022,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00026001713030375664,
|
|
"loss": 4.7404,
|
|
"mean_token_accuracy": 0.1955941289663315,
|
|
"num_tokens": 12477774.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.219074678421021,
|
|
"epoch": 5.695915279878971,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00025962136849759564,
|
|
"loss": 4.7717,
|
|
"mean_token_accuracy": 0.1881052240729332,
|
|
"num_tokens": 12489621.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.236097717285157,
|
|
"epoch": 5.70095814422592,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00025922565448857605,
|
|
"loss": 4.7933,
|
|
"mean_token_accuracy": 0.18855362236499787,
|
|
"num_tokens": 12501885.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.284992313385009,
|
|
"epoch": 5.706001008572869,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00025882998950658623,
|
|
"loss": 4.8248,
|
|
"mean_token_accuracy": 0.1912318855524063,
|
|
"num_tokens": 12514066.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.142426633834839,
|
|
"epoch": 5.711043872919818,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00025843437478136196,
|
|
"loss": 4.7187,
|
|
"mean_token_accuracy": 0.19200052320957184,
|
|
"num_tokens": 12524603.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.1486999034881595,
|
|
"epoch": 5.716086737266767,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00025803881154248305,
|
|
"loss": 4.7089,
|
|
"mean_token_accuracy": 0.20347580909729004,
|
|
"num_tokens": 12536339.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.271569156646729,
|
|
"epoch": 5.7211296016137165,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002576433010193691,
|
|
"loss": 4.7593,
|
|
"mean_token_accuracy": 0.19659196734428405,
|
|
"num_tokens": 12547346.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.1716962337493895,
|
|
"epoch": 5.726172465960666,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000257247844441276,
|
|
"loss": 4.7225,
|
|
"mean_token_accuracy": 0.1943574994802475,
|
|
"num_tokens": 12558064.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.141001510620117,
|
|
"epoch": 5.731215330307615,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00025685244303729187,
|
|
"loss": 4.7146,
|
|
"mean_token_accuracy": 0.19927516281604768,
|
|
"num_tokens": 12568908.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.192882871627807,
|
|
"epoch": 5.736258194654564,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002564570980363334,
|
|
"loss": 4.7675,
|
|
"mean_token_accuracy": 0.18903303146362305,
|
|
"num_tokens": 12579731.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.19461841583252,
|
|
"epoch": 5.741301059001513,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00025606181066714195,
|
|
"loss": 4.7114,
|
|
"mean_token_accuracy": 0.20196596384048462,
|
|
"num_tokens": 12589438.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.1972002506256105,
|
|
"epoch": 5.746343923348462,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002556665821582798,
|
|
"loss": 4.8305,
|
|
"mean_token_accuracy": 0.18280127942562102,
|
|
"num_tokens": 12600204.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.171612882614136,
|
|
"epoch": 5.751386787695411,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002552714137381264,
|
|
"loss": 4.7247,
|
|
"mean_token_accuracy": 0.20713772922754287,
|
|
"num_tokens": 12609441.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.231828784942627,
|
|
"epoch": 5.75642965204236,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002548763066348741,
|
|
"loss": 4.676,
|
|
"mean_token_accuracy": 0.1995038330554962,
|
|
"num_tokens": 12620286.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.219195699691772,
|
|
"epoch": 5.7614725163893095,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002544812620765251,
|
|
"loss": 4.72,
|
|
"mean_token_accuracy": 0.19732799082994462,
|
|
"num_tokens": 12631377.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.231930494308472,
|
|
"epoch": 5.766515380736259,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002540862812908868,
|
|
"loss": 4.7975,
|
|
"mean_token_accuracy": 0.19319574385881425,
|
|
"num_tokens": 12643550.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.185274744033814,
|
|
"epoch": 5.771558245083208,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002536913655055687,
|
|
"loss": 4.7753,
|
|
"mean_token_accuracy": 0.19671137630939484,
|
|
"num_tokens": 12654269.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.224550724029541,
|
|
"epoch": 5.776601109430157,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002532965159479781,
|
|
"loss": 4.7728,
|
|
"mean_token_accuracy": 0.1953396677970886,
|
|
"num_tokens": 12665146.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.218366622924805,
|
|
"epoch": 5.781643973777105,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002529017338453167,
|
|
"loss": 4.7952,
|
|
"mean_token_accuracy": 0.19341515898704528,
|
|
"num_tokens": 12675711.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.220849466323853,
|
|
"epoch": 5.786686838124054,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002525070204245761,
|
|
"loss": 4.6525,
|
|
"mean_token_accuracy": 0.19890592098236085,
|
|
"num_tokens": 12686065.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.161897945404053,
|
|
"epoch": 5.791729702471003,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00025211237691253507,
|
|
"loss": 4.7277,
|
|
"mean_token_accuracy": 0.19724554419517518,
|
|
"num_tokens": 12696581.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.1783380031585695,
|
|
"epoch": 5.796772566817952,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00025171780453575447,
|
|
"loss": 4.7944,
|
|
"mean_token_accuracy": 0.19177568703889847,
|
|
"num_tokens": 12707696.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.19844880104065,
|
|
"epoch": 5.8018154311649015,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002513233045205746,
|
|
"loss": 4.7278,
|
|
"mean_token_accuracy": 0.19615655690431594,
|
|
"num_tokens": 12717992.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.239669752120972,
|
|
"epoch": 5.806858295511851,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002509288780931102,
|
|
"loss": 4.7691,
|
|
"mean_token_accuracy": 0.18855569660663604,
|
|
"num_tokens": 12729309.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.225535488128662,
|
|
"epoch": 5.8119011598588,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002505345264792482,
|
|
"loss": 4.75,
|
|
"mean_token_accuracy": 0.19736499935388566,
|
|
"num_tokens": 12738493.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.170462417602539,
|
|
"epoch": 5.816944024205749,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00025014025090464205,
|
|
"loss": 4.7442,
|
|
"mean_token_accuracy": 0.19254893213510513,
|
|
"num_tokens": 12748797.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.196257781982422,
|
|
"epoch": 5.821986888552698,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00024974605259470975,
|
|
"loss": 4.7065,
|
|
"mean_token_accuracy": 0.19535991549491882,
|
|
"num_tokens": 12760040.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.205750894546509,
|
|
"epoch": 5.827029752899647,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00024935193277462835,
|
|
"loss": 4.7426,
|
|
"mean_token_accuracy": 0.19353136867284776,
|
|
"num_tokens": 12770285.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.187395858764648,
|
|
"epoch": 5.832072617246596,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002489578926693317,
|
|
"loss": 4.7567,
|
|
"mean_token_accuracy": 0.19785541146993638,
|
|
"num_tokens": 12783012.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.232307767868042,
|
|
"epoch": 5.837115481593545,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002485639335035053,
|
|
"loss": 4.8038,
|
|
"mean_token_accuracy": 0.18892906755208969,
|
|
"num_tokens": 12794850.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.255090427398682,
|
|
"epoch": 5.842158345940494,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00024817005650158345,
|
|
"loss": 4.7355,
|
|
"mean_token_accuracy": 0.19619698375463485,
|
|
"num_tokens": 12805087.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.2606336116790775,
|
|
"epoch": 5.8472012102874436,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00024777626288774473,
|
|
"loss": 4.8467,
|
|
"mean_token_accuracy": 0.18575032204389572,
|
|
"num_tokens": 12816144.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.242085599899292,
|
|
"epoch": 5.852244074634393,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002473825538859091,
|
|
"loss": 4.7134,
|
|
"mean_token_accuracy": 0.19740552306175232,
|
|
"num_tokens": 12826942.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.2210101127624515,
|
|
"epoch": 5.857286938981342,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002469889307197331,
|
|
"loss": 4.7283,
|
|
"mean_token_accuracy": 0.19688309580087662,
|
|
"num_tokens": 12836777.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.199464797973633,
|
|
"epoch": 5.86232980332829,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002465953946126065,
|
|
"loss": 4.7701,
|
|
"mean_token_accuracy": 0.19114021062850953,
|
|
"num_tokens": 12848286.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.187376976013184,
|
|
"epoch": 5.867372667675239,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00024620194678764875,
|
|
"loss": 4.6901,
|
|
"mean_token_accuracy": 0.19944073557853698,
|
|
"num_tokens": 12859383.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.169087553024292,
|
|
"epoch": 5.872415532022188,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002458085884677046,
|
|
"loss": 4.7542,
|
|
"mean_token_accuracy": 0.19495588690042495,
|
|
"num_tokens": 12869897.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.236774826049805,
|
|
"epoch": 5.877458396369137,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00024541532087534097,
|
|
"loss": 4.7838,
|
|
"mean_token_accuracy": 0.19512300193309784,
|
|
"num_tokens": 12880052.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.264737224578857,
|
|
"epoch": 5.8825012607160865,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002450221452328424,
|
|
"loss": 4.8547,
|
|
"mean_token_accuracy": 0.18819346874952317,
|
|
"num_tokens": 12892322.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.203943586349487,
|
|
"epoch": 5.887544125063036,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00024462906276220814,
|
|
"loss": 4.7424,
|
|
"mean_token_accuracy": 0.1953052505850792,
|
|
"num_tokens": 12903399.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.169946479797363,
|
|
"epoch": 5.892586989409985,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002442360746851473,
|
|
"loss": 4.6944,
|
|
"mean_token_accuracy": 0.19720448404550553,
|
|
"num_tokens": 12913108.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.245753049850464,
|
|
"epoch": 5.897629853756934,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00024384318222307605,
|
|
"loss": 4.7955,
|
|
"mean_token_accuracy": 0.19111020117998123,
|
|
"num_tokens": 12924810.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.26667890548706,
|
|
"epoch": 5.902672718103883,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002434503865971131,
|
|
"loss": 4.7995,
|
|
"mean_token_accuracy": 0.19105137139558792,
|
|
"num_tokens": 12934896.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.186099338531494,
|
|
"epoch": 5.907715582450832,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002430576890280765,
|
|
"loss": 4.7382,
|
|
"mean_token_accuracy": 0.19835858941078185,
|
|
"num_tokens": 12945873.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.210193395614624,
|
|
"epoch": 5.912758446797781,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00024266509073647913,
|
|
"loss": 4.8497,
|
|
"mean_token_accuracy": 0.18862408697605132,
|
|
"num_tokens": 12958701.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.200048971176147,
|
|
"epoch": 5.91780131114473,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00024227259294252578,
|
|
"loss": 4.8079,
|
|
"mean_token_accuracy": 0.1938364699482918,
|
|
"num_tokens": 12970729.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.217307424545288,
|
|
"epoch": 5.922844175491679,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002418801968661084,
|
|
"loss": 4.7664,
|
|
"mean_token_accuracy": 0.19207355082035066,
|
|
"num_tokens": 12980917.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.2100639820098875,
|
|
"epoch": 5.9278870398386285,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00024148790372680324,
|
|
"loss": 4.8131,
|
|
"mean_token_accuracy": 0.19460143148899078,
|
|
"num_tokens": 12993123.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.1463652610778805,
|
|
"epoch": 5.932929904185578,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00024109571474386631,
|
|
"loss": 4.6353,
|
|
"mean_token_accuracy": 0.21409369856119156,
|
|
"num_tokens": 13004031.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.250701189041138,
|
|
"epoch": 5.937972768532527,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00024070363113623018,
|
|
"loss": 4.827,
|
|
"mean_token_accuracy": 0.19517660290002822,
|
|
"num_tokens": 13015407.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.218714761734009,
|
|
"epoch": 5.943015632879476,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00024031165412249938,
|
|
"loss": 4.7433,
|
|
"mean_token_accuracy": 0.20145001411437988,
|
|
"num_tokens": 13027293.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.242775869369507,
|
|
"epoch": 5.948058497226425,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000239919784920948,
|
|
"loss": 4.7555,
|
|
"mean_token_accuracy": 0.19821065813302993,
|
|
"num_tokens": 13037033.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.233956670761108,
|
|
"epoch": 5.953101361573374,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00023952802474951418,
|
|
"loss": 4.7874,
|
|
"mean_token_accuracy": 0.19476954489946366,
|
|
"num_tokens": 13049896.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.177340126037597,
|
|
"epoch": 5.958144225920323,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00023913637482579774,
|
|
"loss": 4.7632,
|
|
"mean_token_accuracy": 0.1902133896946907,
|
|
"num_tokens": 13059745.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.2015830993652346,
|
|
"epoch": 5.9631870902672715,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00023874483636705563,
|
|
"loss": 4.7629,
|
|
"mean_token_accuracy": 0.19815345704555512,
|
|
"num_tokens": 13070993.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.28646445274353,
|
|
"epoch": 5.968229954614221,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00023835341059019838,
|
|
"loss": 4.8996,
|
|
"mean_token_accuracy": 0.18579458743333815,
|
|
"num_tokens": 13081743.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.262183761596679,
|
|
"epoch": 5.97327281896117,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00023796209871178642,
|
|
"loss": 4.7768,
|
|
"mean_token_accuracy": 0.1921431988477707,
|
|
"num_tokens": 13092894.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.176323461532593,
|
|
"epoch": 5.978315683308119,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00023757090194802605,
|
|
"loss": 4.7076,
|
|
"mean_token_accuracy": 0.20235335528850557,
|
|
"num_tokens": 13103886.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.235933303833008,
|
|
"epoch": 5.983358547655068,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00023717982151476592,
|
|
"loss": 4.8153,
|
|
"mean_token_accuracy": 0.18534866571426392,
|
|
"num_tokens": 13114858.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.255216455459594,
|
|
"epoch": 5.988401412002017,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00023678885862749282,
|
|
"loss": 4.8137,
|
|
"mean_token_accuracy": 0.1926460474729538,
|
|
"num_tokens": 13126359.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.163140821456909,
|
|
"epoch": 5.993444276348966,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00023639801450132863,
|
|
"loss": 4.6684,
|
|
"mean_token_accuracy": 0.1935342401266098,
|
|
"num_tokens": 13136623.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.198820352554321,
|
|
"epoch": 5.998487140695915,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00023600729035102585,
|
|
"loss": 4.7816,
|
|
"mean_token_accuracy": 0.19893045723438263,
|
|
"num_tokens": 13147518.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.214005364312066,
|
|
"epoch": 6.0030257186081695,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002356166873909642,
|
|
"loss": 4.5889,
|
|
"mean_token_accuracy": 0.20765161845419142,
|
|
"num_tokens": 13155773.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.225727605819702,
|
|
"epoch": 6.008068582955119,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00023522620683514645,
|
|
"loss": 4.651,
|
|
"mean_token_accuracy": 0.2064117968082428,
|
|
"num_tokens": 13167682.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.195204496383667,
|
|
"epoch": 6.013111447302068,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00023483584989719553,
|
|
"loss": 4.5298,
|
|
"mean_token_accuracy": 0.2180487707257271,
|
|
"num_tokens": 13179069.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.23414626121521,
|
|
"epoch": 6.018154311649017,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00023444561779034957,
|
|
"loss": 4.5559,
|
|
"mean_token_accuracy": 0.20881845206022262,
|
|
"num_tokens": 13190548.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.16019458770752,
|
|
"epoch": 6.023197175995966,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00023405551172745904,
|
|
"loss": 4.5512,
|
|
"mean_token_accuracy": 0.21359633803367614,
|
|
"num_tokens": 13202006.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 5.217672204971313,
|
|
"epoch": 6.028240040342915,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00023366553292098247,
|
|
"loss": 4.5698,
|
|
"mean_token_accuracy": 0.2145567297935486,
|
|
"num_tokens": 13212236.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.184329557418823,
|
|
"epoch": 6.033282904689864,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002332756825829832,
|
|
"loss": 4.5477,
|
|
"mean_token_accuracy": 0.20363112688064575,
|
|
"num_tokens": 13224704.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.151283073425293,
|
|
"epoch": 6.038325769036813,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002328859619251249,
|
|
"loss": 4.5029,
|
|
"mean_token_accuracy": 0.2145952805876732,
|
|
"num_tokens": 13236020.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.15667085647583,
|
|
"epoch": 6.043368633383762,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00023249637215866847,
|
|
"loss": 4.5586,
|
|
"mean_token_accuracy": 0.20535304248332978,
|
|
"num_tokens": 13247466.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.174594974517822,
|
|
"epoch": 6.048411497730711,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002321069144944678,
|
|
"loss": 4.5655,
|
|
"mean_token_accuracy": 0.21546992659568787,
|
|
"num_tokens": 13258927.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 6.048411497730711,
|
|
"eval_entropy": 5.153881462475725,
|
|
"eval_loss": 5.7050347328186035,
|
|
"eval_mean_token_accuracy": 0.15401927290754464,
|
|
"eval_num_tokens": 13258927.0,
|
|
"eval_runtime": 2.651,
|
|
"eval_samples_per_second": 1183.707,
|
|
"eval_steps_per_second": 148.246,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 5.260726594924927,
|
|
"epoch": 6.05345436207766,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00023171759014296639,
|
|
"loss": 4.5778,
|
|
"mean_token_accuracy": 0.20787244141101838,
|
|
"num_tokens": 13269063.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 5.158452796936035,
|
|
"epoch": 6.058497226424609,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00023132840031419316,
|
|
"loss": 4.5503,
|
|
"mean_token_accuracy": 0.21846787333488465,
|
|
"num_tokens": 13280199.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 5.206730079650879,
|
|
"epoch": 6.063540090771558,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00023093934621775926,
|
|
"loss": 4.6033,
|
|
"mean_token_accuracy": 0.20758186429738998,
|
|
"num_tokens": 13291324.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 5.192326974868775,
|
|
"epoch": 6.068582955118507,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00023055042906285368,
|
|
"loss": 4.6443,
|
|
"mean_token_accuracy": 0.20416068881750107,
|
|
"num_tokens": 13303833.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 5.237275266647339,
|
|
"epoch": 6.073625819465456,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002301616500582398,
|
|
"loss": 4.5903,
|
|
"mean_token_accuracy": 0.21257395893335343,
|
|
"num_tokens": 13314806.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 5.242798757553101,
|
|
"epoch": 6.078668683812405,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00022977301041225207,
|
|
"loss": 4.6493,
|
|
"mean_token_accuracy": 0.20611777901649475,
|
|
"num_tokens": 13326133.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 5.159042835235596,
|
|
"epoch": 6.083711548159354,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00022938451133279114,
|
|
"loss": 4.4898,
|
|
"mean_token_accuracy": 0.21559949964284897,
|
|
"num_tokens": 13337957.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 5.24175124168396,
|
|
"epoch": 6.088754412506304,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00022899615402732133,
|
|
"loss": 4.5752,
|
|
"mean_token_accuracy": 0.2092887967824936,
|
|
"num_tokens": 13348242.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 5.22891526222229,
|
|
"epoch": 6.093797276853253,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00022860793970286592,
|
|
"loss": 4.6443,
|
|
"mean_token_accuracy": 0.20402273684740066,
|
|
"num_tokens": 13360474.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 5.188431882858277,
|
|
"epoch": 6.098840141200202,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002282198695660042,
|
|
"loss": 4.6466,
|
|
"mean_token_accuracy": 0.20813622027635575,
|
|
"num_tokens": 13371616.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 5.219429063796997,
|
|
"epoch": 6.103883005547151,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00022783194482286684,
|
|
"loss": 4.5588,
|
|
"mean_token_accuracy": 0.21435497254133223,
|
|
"num_tokens": 13381297.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 5.191244316101074,
|
|
"epoch": 6.1089258698941,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002274441666791332,
|
|
"loss": 4.5996,
|
|
"mean_token_accuracy": 0.20570364743471145,
|
|
"num_tokens": 13391847.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 5.107387971878052,
|
|
"epoch": 6.113968734241049,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00022705653634002648,
|
|
"loss": 4.4593,
|
|
"mean_token_accuracy": 0.2136998549103737,
|
|
"num_tokens": 13403784.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 5.102633619308472,
|
|
"epoch": 6.119011598587998,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00022666905501031077,
|
|
"loss": 4.5584,
|
|
"mean_token_accuracy": 0.20832539051771165,
|
|
"num_tokens": 13414547.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 5.169452810287476,
|
|
"epoch": 6.124054462934947,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002262817238942869,
|
|
"loss": 4.532,
|
|
"mean_token_accuracy": 0.20973353683948517,
|
|
"num_tokens": 13423571.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 5.226801347732544,
|
|
"epoch": 6.1290973272818965,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002258945441957892,
|
|
"loss": 4.6234,
|
|
"mean_token_accuracy": 0.20691546499729158,
|
|
"num_tokens": 13434281.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 5.15013861656189,
|
|
"epoch": 6.134140191628846,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002255075171181807,
|
|
"loss": 4.5144,
|
|
"mean_token_accuracy": 0.21510060727596284,
|
|
"num_tokens": 13444795.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 5.212281513214111,
|
|
"epoch": 6.139183055975794,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00022512064386435082,
|
|
"loss": 4.6699,
|
|
"mean_token_accuracy": 0.2050182431936264,
|
|
"num_tokens": 13456671.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 5.227065467834473,
|
|
"epoch": 6.144225920322743,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002247339256367103,
|
|
"loss": 4.5811,
|
|
"mean_token_accuracy": 0.20647963434457778,
|
|
"num_tokens": 13466981.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 5.216949462890625,
|
|
"epoch": 6.149268784669692,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00022434736363718854,
|
|
"loss": 4.6433,
|
|
"mean_token_accuracy": 0.20096343159675598,
|
|
"num_tokens": 13479650.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 5.204792213439942,
|
|
"epoch": 6.154311649016641,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00022396095906722897,
|
|
"loss": 4.5997,
|
|
"mean_token_accuracy": 0.20954020768404008,
|
|
"num_tokens": 13491339.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 5.203112506866455,
|
|
"epoch": 6.15935451336359,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002235747131277861,
|
|
"loss": 4.5178,
|
|
"mean_token_accuracy": 0.21020906567573547,
|
|
"num_tokens": 13501634.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 5.118396949768067,
|
|
"epoch": 6.164397377710539,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00022318862701932102,
|
|
"loss": 4.52,
|
|
"mean_token_accuracy": 0.210641485452652,
|
|
"num_tokens": 13511372.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 5.105551290512085,
|
|
"epoch": 6.1694402420574885,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00022280270194179836,
|
|
"loss": 4.4959,
|
|
"mean_token_accuracy": 0.2211366519331932,
|
|
"num_tokens": 13522872.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 5.160684967041016,
|
|
"epoch": 6.174483106404438,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00022241693909468214,
|
|
"loss": 4.5307,
|
|
"mean_token_accuracy": 0.21257291585206986,
|
|
"num_tokens": 13534719.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 5.191540098190307,
|
|
"epoch": 6.179525970751387,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00022203133967693217,
|
|
"loss": 4.6257,
|
|
"mean_token_accuracy": 0.20276657938957215,
|
|
"num_tokens": 13545763.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 5.217547512054443,
|
|
"epoch": 6.184568835098336,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00022164590488700042,
|
|
"loss": 4.538,
|
|
"mean_token_accuracy": 0.20638973116874695,
|
|
"num_tokens": 13556177.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 5.150887584686279,
|
|
"epoch": 6.189611699445285,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00022126063592282696,
|
|
"loss": 4.5589,
|
|
"mean_token_accuracy": 0.2077105775475502,
|
|
"num_tokens": 13567258.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 5.175705289840698,
|
|
"epoch": 6.194654563792234,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00022087553398183684,
|
|
"loss": 4.5692,
|
|
"mean_token_accuracy": 0.21554771512746812,
|
|
"num_tokens": 13577118.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 5.136524200439453,
|
|
"epoch": 6.199697428139183,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00022049060026093555,
|
|
"loss": 4.54,
|
|
"mean_token_accuracy": 0.21474739164113998,
|
|
"num_tokens": 13589123.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 5.170968008041382,
|
|
"epoch": 6.204740292486132,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00022010583595650617,
|
|
"loss": 4.5159,
|
|
"mean_token_accuracy": 0.2132031112909317,
|
|
"num_tokens": 13601080.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 5.182331085205078,
|
|
"epoch": 6.2097831568330815,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00021972124226440483,
|
|
"loss": 4.555,
|
|
"mean_token_accuracy": 0.21134361028671264,
|
|
"num_tokens": 13612779.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 5.154096889495849,
|
|
"epoch": 6.214826021180031,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00021933682037995772,
|
|
"loss": 4.564,
|
|
"mean_token_accuracy": 0.21000760942697524,
|
|
"num_tokens": 13623659.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 5.21182918548584,
|
|
"epoch": 6.21986888552698,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00021895257149795694,
|
|
"loss": 4.6225,
|
|
"mean_token_accuracy": 0.20887086689472198,
|
|
"num_tokens": 13634124.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 5.159660720825196,
|
|
"epoch": 6.224911749873929,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002185684968126569,
|
|
"loss": 4.5342,
|
|
"mean_token_accuracy": 0.21438640356063843,
|
|
"num_tokens": 13645190.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 5.183885955810547,
|
|
"epoch": 6.229954614220877,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002181845975177703,
|
|
"loss": 4.5535,
|
|
"mean_token_accuracy": 0.21194448173046113,
|
|
"num_tokens": 13655895.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 5.137746953964234,
|
|
"epoch": 6.234997478567826,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00021780087480646548,
|
|
"loss": 4.4716,
|
|
"mean_token_accuracy": 0.21419171541929244,
|
|
"num_tokens": 13666304.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 5.1655854225158695,
|
|
"epoch": 6.240040342914775,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002174173298713611,
|
|
"loss": 4.5559,
|
|
"mean_token_accuracy": 0.21139683425426484,
|
|
"num_tokens": 13676574.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 5.130858564376831,
|
|
"epoch": 6.245083207261724,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00021703396390452385,
|
|
"loss": 4.5345,
|
|
"mean_token_accuracy": 0.21129630208015443,
|
|
"num_tokens": 13686985.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 5.1676350116729735,
|
|
"epoch": 6.2501260716086735,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002166507780974639,
|
|
"loss": 4.6391,
|
|
"mean_token_accuracy": 0.21014945805072785,
|
|
"num_tokens": 13697058.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 5.172171020507813,
|
|
"epoch": 6.255168935955623,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002162677736411317,
|
|
"loss": 4.6182,
|
|
"mean_token_accuracy": 0.19714466631412506,
|
|
"num_tokens": 13708576.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 5.213000202178955,
|
|
"epoch": 6.260211800302572,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00021588495172591377,
|
|
"loss": 4.6483,
|
|
"mean_token_accuracy": 0.20235108584165573,
|
|
"num_tokens": 13719792.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 5.248628759384156,
|
|
"epoch": 6.265254664649521,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00021550231354162957,
|
|
"loss": 4.644,
|
|
"mean_token_accuracy": 0.20657316595315933,
|
|
"num_tokens": 13730147.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 5.170094728469849,
|
|
"epoch": 6.27029752899647,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021511986027752733,
|
|
"loss": 4.553,
|
|
"mean_token_accuracy": 0.20549349635839462,
|
|
"num_tokens": 13741559.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 5.194694232940674,
|
|
"epoch": 6.275340393343419,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002147375931222807,
|
|
"loss": 4.5722,
|
|
"mean_token_accuracy": 0.20254788994789125,
|
|
"num_tokens": 13753082.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 5.172289133071899,
|
|
"epoch": 6.280383257690368,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00021435551326398478,
|
|
"loss": 4.6495,
|
|
"mean_token_accuracy": 0.20407428443431855,
|
|
"num_tokens": 13764475.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 5.153161144256591,
|
|
"epoch": 6.285426122037317,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00021397362189015263,
|
|
"loss": 4.6391,
|
|
"mean_token_accuracy": 0.21152440905570985,
|
|
"num_tokens": 13776411.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 5.1737635135650635,
|
|
"epoch": 6.290468986384266,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00021359192018771127,
|
|
"loss": 4.5625,
|
|
"mean_token_accuracy": 0.21435577273368836,
|
|
"num_tokens": 13786443.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 5.1560218811035154,
|
|
"epoch": 6.2955118507312156,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00021321040934299852,
|
|
"loss": 4.5968,
|
|
"mean_token_accuracy": 0.20996685028076173,
|
|
"num_tokens": 13797613.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 5.144155120849609,
|
|
"epoch": 6.300554715078165,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00021282909054175903,
|
|
"loss": 4.5828,
|
|
"mean_token_accuracy": 0.21616074442863464,
|
|
"num_tokens": 13809575.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 5.126867341995239,
|
|
"epoch": 6.305597579425114,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002124479649691402,
|
|
"loss": 4.5247,
|
|
"mean_token_accuracy": 0.21688440442085266,
|
|
"num_tokens": 13821009.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 5.195837831497192,
|
|
"epoch": 6.310640443772063,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00021206703380968923,
|
|
"loss": 4.5755,
|
|
"mean_token_accuracy": 0.20989220589399338,
|
|
"num_tokens": 13831659.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 5.198780107498169,
|
|
"epoch": 6.315683308119011,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00021168629824734887,
|
|
"loss": 4.5895,
|
|
"mean_token_accuracy": 0.2107818305492401,
|
|
"num_tokens": 13842489.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 5.153617286682129,
|
|
"epoch": 6.32072617246596,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00021130575946545416,
|
|
"loss": 4.5808,
|
|
"mean_token_accuracy": 0.2079627811908722,
|
|
"num_tokens": 13854295.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 5.1764017105102536,
|
|
"epoch": 6.325769036812909,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00021092541864672825,
|
|
"loss": 4.5227,
|
|
"mean_token_accuracy": 0.22012332528829576,
|
|
"num_tokens": 13864761.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 5.146976375579834,
|
|
"epoch": 6.3308119011598585,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021054527697327935,
|
|
"loss": 4.581,
|
|
"mean_token_accuracy": 0.2136762708425522,
|
|
"num_tokens": 13875937.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 5.12059907913208,
|
|
"epoch": 6.335854765506808,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00021016533562659634,
|
|
"loss": 4.4902,
|
|
"mean_token_accuracy": 0.21374871581792831,
|
|
"num_tokens": 13886114.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 5.15385594367981,
|
|
"epoch": 6.340897629853757,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00020978559578754586,
|
|
"loss": 4.5906,
|
|
"mean_token_accuracy": 0.21113584637641908,
|
|
"num_tokens": 13896542.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 5.197323513031006,
|
|
"epoch": 6.345940494200706,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000209406058636368,
|
|
"loss": 4.6342,
|
|
"mean_token_accuracy": 0.20339771509170532,
|
|
"num_tokens": 13907235.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 5.255248355865478,
|
|
"epoch": 6.350983358547655,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00020902672535267315,
|
|
"loss": 4.6327,
|
|
"mean_token_accuracy": 0.20695483833551406,
|
|
"num_tokens": 13918405.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 5.144094657897949,
|
|
"epoch": 6.356026222894604,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00020864759711543762,
|
|
"loss": 4.5493,
|
|
"mean_token_accuracy": 0.21040454357862473,
|
|
"num_tokens": 13929786.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 5.123776197433472,
|
|
"epoch": 6.361069087241553,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002082686751030009,
|
|
"loss": 4.5759,
|
|
"mean_token_accuracy": 0.2096346691250801,
|
|
"num_tokens": 13940801.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 5.255251550674439,
|
|
"epoch": 6.366111951588502,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002078899604930613,
|
|
"loss": 4.7304,
|
|
"mean_token_accuracy": 0.20091987848281861,
|
|
"num_tokens": 13952582.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 5.186233186721802,
|
|
"epoch": 6.371154815935451,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00020751145446267267,
|
|
"loss": 4.6158,
|
|
"mean_token_accuracy": 0.2028737723827362,
|
|
"num_tokens": 13963741.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 5.231812858581543,
|
|
"epoch": 6.3761976802824005,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002071331581882403,
|
|
"loss": 4.5977,
|
|
"mean_token_accuracy": 0.20814272314310073,
|
|
"num_tokens": 13975024.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 5.1858665466308596,
|
|
"epoch": 6.38124054462935,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020675507284551793,
|
|
"loss": 4.5606,
|
|
"mean_token_accuracy": 0.21160346418619155,
|
|
"num_tokens": 13984735.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 5.191213226318359,
|
|
"epoch": 6.386283408976299,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00020637719960960333,
|
|
"loss": 4.5944,
|
|
"mean_token_accuracy": 0.20684802979230882,
|
|
"num_tokens": 13996526.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 5.2244853496551515,
|
|
"epoch": 6.391326273323248,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020599953965493535,
|
|
"loss": 4.6352,
|
|
"mean_token_accuracy": 0.21371518671512604,
|
|
"num_tokens": 14008357.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 5.137908840179444,
|
|
"epoch": 6.396369137670197,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00020562209415528982,
|
|
"loss": 4.6063,
|
|
"mean_token_accuracy": 0.2148994266986847,
|
|
"num_tokens": 14019973.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 5.156728839874267,
|
|
"epoch": 6.401412002017146,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00020524486428377605,
|
|
"loss": 4.5916,
|
|
"mean_token_accuracy": 0.209877148270607,
|
|
"num_tokens": 14032091.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 5.231080198287964,
|
|
"epoch": 6.406454866364095,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002048678512128332,
|
|
"loss": 4.6077,
|
|
"mean_token_accuracy": 0.2056787982583046,
|
|
"num_tokens": 14043698.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 5.147547054290771,
|
|
"epoch": 6.411497730711044,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00020449105611422658,
|
|
"loss": 4.551,
|
|
"mean_token_accuracy": 0.21446956396102906,
|
|
"num_tokens": 14055041.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 5.1980408191680905,
|
|
"epoch": 6.416540595057993,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00020411448015904415,
|
|
"loss": 4.6106,
|
|
"mean_token_accuracy": 0.20289706587791442,
|
|
"num_tokens": 14066094.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 5.180946922302246,
|
|
"epoch": 6.421583459404942,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00020373812451769248,
|
|
"loss": 4.572,
|
|
"mean_token_accuracy": 0.21222969889640808,
|
|
"num_tokens": 14075851.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 5.189320135116577,
|
|
"epoch": 6.426626323751891,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00020336199035989372,
|
|
"loss": 4.5706,
|
|
"mean_token_accuracy": 0.20867748856544494,
|
|
"num_tokens": 14087204.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 5.138676452636719,
|
|
"epoch": 6.43166918809884,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002029860788546814,
|
|
"loss": 4.6223,
|
|
"mean_token_accuracy": 0.20294176340103148,
|
|
"num_tokens": 14097473.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 5.218420553207397,
|
|
"epoch": 6.436712052445789,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00020261039117039725,
|
|
"loss": 4.6469,
|
|
"mean_token_accuracy": 0.20407562851905822,
|
|
"num_tokens": 14109469.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 5.20270357131958,
|
|
"epoch": 6.441754916792738,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00020223492847468722,
|
|
"loss": 4.6132,
|
|
"mean_token_accuracy": 0.21009938716888427,
|
|
"num_tokens": 14119883.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 5.138283205032349,
|
|
"epoch": 6.446797781139687,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002018596919344981,
|
|
"loss": 4.5331,
|
|
"mean_token_accuracy": 0.20844844877719879,
|
|
"num_tokens": 14131287.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 5.136653232574463,
|
|
"epoch": 6.451840645486636,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00020148468271607352,
|
|
"loss": 4.5767,
|
|
"mean_token_accuracy": 0.21220425069332122,
|
|
"num_tokens": 14142926.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 5.229744911193848,
|
|
"epoch": 6.4568835098335855,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00020110990198495105,
|
|
"loss": 4.685,
|
|
"mean_token_accuracy": 0.20225642174482344,
|
|
"num_tokens": 14153919.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 5.1576436996459964,
|
|
"epoch": 6.461926374180535,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002007353509059575,
|
|
"loss": 4.5977,
|
|
"mean_token_accuracy": 0.20400530248880386,
|
|
"num_tokens": 14165246.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 5.165514421463013,
|
|
"epoch": 6.466969238527484,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00020036103064320664,
|
|
"loss": 4.6244,
|
|
"mean_token_accuracy": 0.1998355507850647,
|
|
"num_tokens": 14176274.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 5.194111061096192,
|
|
"epoch": 6.472012102874433,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00019998694236009423,
|
|
"loss": 4.6362,
|
|
"mean_token_accuracy": 0.20124897211790085,
|
|
"num_tokens": 14187053.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 5.226767444610596,
|
|
"epoch": 6.477054967221382,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001996130872192955,
|
|
"loss": 4.6416,
|
|
"mean_token_accuracy": 0.19850814044475557,
|
|
"num_tokens": 14198737.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 5.198610305786133,
|
|
"epoch": 6.482097831568331,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00019923946638276052,
|
|
"loss": 4.5954,
|
|
"mean_token_accuracy": 0.20362408608198165,
|
|
"num_tokens": 14208846.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 5.166561841964722,
|
|
"epoch": 6.48714069591528,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001988660810117118,
|
|
"loss": 4.5619,
|
|
"mean_token_accuracy": 0.20903020799160005,
|
|
"num_tokens": 14221097.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 5.184852266311646,
|
|
"epoch": 6.492183560262229,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001984929322666395,
|
|
"loss": 4.5594,
|
|
"mean_token_accuracy": 0.21151221990585328,
|
|
"num_tokens": 14232987.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 5.176004314422608,
|
|
"epoch": 6.4972264246091775,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00019812002130729857,
|
|
"loss": 4.5487,
|
|
"mean_token_accuracy": 0.21392588764429094,
|
|
"num_tokens": 14244309.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 5.152710199356079,
|
|
"epoch": 6.502269288956127,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00019774734929270483,
|
|
"loss": 4.6211,
|
|
"mean_token_accuracy": 0.20146573334932327,
|
|
"num_tokens": 14255619.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 5.124300384521485,
|
|
"epoch": 6.507312153303076,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00019737491738113167,
|
|
"loss": 4.5238,
|
|
"mean_token_accuracy": 0.2194095641374588,
|
|
"num_tokens": 14266115.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 5.174131965637207,
|
|
"epoch": 6.512355017650025,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00019700272673010588,
|
|
"loss": 4.6379,
|
|
"mean_token_accuracy": 0.2015240266919136,
|
|
"num_tokens": 14276936.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 5.185207986831665,
|
|
"epoch": 6.517397881996974,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001966307784964046,
|
|
"loss": 4.6335,
|
|
"mean_token_accuracy": 0.19825737923383713,
|
|
"num_tokens": 14288350.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 5.188508653640747,
|
|
"epoch": 6.522440746343923,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00019625907383605162,
|
|
"loss": 4.6468,
|
|
"mean_token_accuracy": 0.20679865330457686,
|
|
"num_tokens": 14301269.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 5.223599290847778,
|
|
"epoch": 6.527483610690872,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00019588761390431354,
|
|
"loss": 4.6175,
|
|
"mean_token_accuracy": 0.2026480957865715,
|
|
"num_tokens": 14312216.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 5.139459228515625,
|
|
"epoch": 6.532526475037821,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00019551639985569642,
|
|
"loss": 4.6058,
|
|
"mean_token_accuracy": 0.2067226454615593,
|
|
"num_tokens": 14324705.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 5.2005781650543215,
|
|
"epoch": 6.5375693393847705,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00019514543284394203,
|
|
"loss": 4.6085,
|
|
"mean_token_accuracy": 0.20172022879123688,
|
|
"num_tokens": 14335009.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 5.182702159881591,
|
|
"epoch": 6.54261220373172,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00019477471402202452,
|
|
"loss": 4.6114,
|
|
"mean_token_accuracy": 0.21173812747001647,
|
|
"num_tokens": 14345881.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 5.158409357070923,
|
|
"epoch": 6.547655068078669,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00019440424454214645,
|
|
"loss": 4.6106,
|
|
"mean_token_accuracy": 0.2060550794005394,
|
|
"num_tokens": 14356175.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 5.141512441635132,
|
|
"epoch": 6.552697932425618,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00019403402555573547,
|
|
"loss": 4.5716,
|
|
"mean_token_accuracy": 0.2106699213385582,
|
|
"num_tokens": 14366480.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"epoch": 6.552697932425618,
|
|
"eval_entropy": 5.109214495156557,
|
|
"eval_loss": 5.706435680389404,
|
|
"eval_mean_token_accuracy": 0.1541565934968662,
|
|
"eval_num_tokens": 14366480.0,
|
|
"eval_runtime": 2.6585,
|
|
"eval_samples_per_second": 1180.387,
|
|
"eval_steps_per_second": 147.83,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 5.117092514038086,
|
|
"epoch": 6.557740796772567,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00019366405821344074,
|
|
"loss": 4.5685,
|
|
"mean_token_accuracy": 0.21564687192440032,
|
|
"num_tokens": 14378064.0,
|
|
"step": 6505
|
|
},
|
|
{
|
|
"entropy": 5.160234642028809,
|
|
"epoch": 6.562783661119516,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001932943436651295,
|
|
"loss": 4.5637,
|
|
"mean_token_accuracy": 0.20918753892183303,
|
|
"num_tokens": 14388860.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 5.27758469581604,
|
|
"epoch": 6.567826525466465,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00019292488305988276,
|
|
"loss": 4.6755,
|
|
"mean_token_accuracy": 0.20482370257377625,
|
|
"num_tokens": 14399359.0,
|
|
"step": 6515
|
|
},
|
|
{
|
|
"entropy": 5.137232875823974,
|
|
"epoch": 6.572869389813414,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001925556775459929,
|
|
"loss": 4.554,
|
|
"mean_token_accuracy": 0.2140171006321907,
|
|
"num_tokens": 14411089.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 5.210144758224487,
|
|
"epoch": 6.577912254160363,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000192186728270959,
|
|
"loss": 4.6884,
|
|
"mean_token_accuracy": 0.20185265094041824,
|
|
"num_tokens": 14422263.0,
|
|
"step": 6525
|
|
},
|
|
{
|
|
"entropy": 5.233591985702515,
|
|
"epoch": 6.5829551185073125,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000191818036381484,
|
|
"loss": 4.6218,
|
|
"mean_token_accuracy": 0.20353255569934844,
|
|
"num_tokens": 14432217.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 5.247843456268311,
|
|
"epoch": 6.587997982854262,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00019144960302347065,
|
|
"loss": 4.6669,
|
|
"mean_token_accuracy": 0.1983882814645767,
|
|
"num_tokens": 14444394.0,
|
|
"step": 6535
|
|
},
|
|
{
|
|
"entropy": 5.1189710140228275,
|
|
"epoch": 6.593040847201211,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00019108142934201854,
|
|
"loss": 4.5685,
|
|
"mean_token_accuracy": 0.22012270241975784,
|
|
"num_tokens": 14456802.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 5.189018821716308,
|
|
"epoch": 6.598083711548159,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00019071351648141977,
|
|
"loss": 4.5975,
|
|
"mean_token_accuracy": 0.20756238996982573,
|
|
"num_tokens": 14467602.0,
|
|
"step": 6545
|
|
},
|
|
{
|
|
"entropy": 5.125365161895752,
|
|
"epoch": 6.603126575895108,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00019034586558515625,
|
|
"loss": 4.4805,
|
|
"mean_token_accuracy": 0.21055586636066437,
|
|
"num_tokens": 14478454.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 5.1714588642120365,
|
|
"epoch": 6.608169440242057,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00018997847779589518,
|
|
"loss": 4.6414,
|
|
"mean_token_accuracy": 0.19963551163673401,
|
|
"num_tokens": 14490108.0,
|
|
"step": 6555
|
|
},
|
|
{
|
|
"entropy": 5.230170822143554,
|
|
"epoch": 6.613212304589006,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00018961135425548642,
|
|
"loss": 4.6464,
|
|
"mean_token_accuracy": 0.20076310336589814,
|
|
"num_tokens": 14500065.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 5.170799398422242,
|
|
"epoch": 6.6182551689359554,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00018924449610495836,
|
|
"loss": 4.5719,
|
|
"mean_token_accuracy": 0.215500171482563,
|
|
"num_tokens": 14509945.0,
|
|
"step": 6565
|
|
},
|
|
{
|
|
"entropy": 5.1722414016723635,
|
|
"epoch": 6.623298033282905,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00018887790448451458,
|
|
"loss": 4.6169,
|
|
"mean_token_accuracy": 0.20690080225467683,
|
|
"num_tokens": 14520980.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 5.15509819984436,
|
|
"epoch": 6.628340897629854,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001885115805335304,
|
|
"loss": 4.5517,
|
|
"mean_token_accuracy": 0.21298984438180923,
|
|
"num_tokens": 14532020.0,
|
|
"step": 6575
|
|
},
|
|
{
|
|
"entropy": 5.151628208160401,
|
|
"epoch": 6.633383761976803,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00018814552539054896,
|
|
"loss": 4.5671,
|
|
"mean_token_accuracy": 0.21397893726825715,
|
|
"num_tokens": 14543313.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 5.197632122039795,
|
|
"epoch": 6.638426626323752,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00018777974019327826,
|
|
"loss": 4.6425,
|
|
"mean_token_accuracy": 0.20667950063943863,
|
|
"num_tokens": 14555397.0,
|
|
"step": 6585
|
|
},
|
|
{
|
|
"entropy": 5.173119831085205,
|
|
"epoch": 6.643469490670701,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00018741422607858693,
|
|
"loss": 4.5194,
|
|
"mean_token_accuracy": 0.21303518116474152,
|
|
"num_tokens": 14564485.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 5.145395088195801,
|
|
"epoch": 6.64851235501765,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00018704898418250143,
|
|
"loss": 4.6073,
|
|
"mean_token_accuracy": 0.19919801503419876,
|
|
"num_tokens": 14575637.0,
|
|
"step": 6595
|
|
},
|
|
{
|
|
"entropy": 5.100871849060058,
|
|
"epoch": 6.653555219364599,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00018668401564020192,
|
|
"loss": 4.5127,
|
|
"mean_token_accuracy": 0.21471484154462814,
|
|
"num_tokens": 14586463.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 5.126468086242676,
|
|
"epoch": 6.658598083711548,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00018631932158601905,
|
|
"loss": 4.5929,
|
|
"mean_token_accuracy": 0.20866924226284028,
|
|
"num_tokens": 14597462.0,
|
|
"step": 6605
|
|
},
|
|
{
|
|
"entropy": 5.14214448928833,
|
|
"epoch": 6.6636409480584975,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00018595490315343038,
|
|
"loss": 4.5367,
|
|
"mean_token_accuracy": 0.21089996248483658,
|
|
"num_tokens": 14608390.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 5.224957513809204,
|
|
"epoch": 6.668683812405447,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00018559076147505688,
|
|
"loss": 4.6531,
|
|
"mean_token_accuracy": 0.20568040162324905,
|
|
"num_tokens": 14620935.0,
|
|
"step": 6615
|
|
},
|
|
{
|
|
"entropy": 5.105131769180298,
|
|
"epoch": 6.673726676752396,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00018522689768265911,
|
|
"loss": 4.5937,
|
|
"mean_token_accuracy": 0.20684238374233246,
|
|
"num_tokens": 14633108.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 5.112280941009521,
|
|
"epoch": 6.678769541099344,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00018486331290713427,
|
|
"loss": 4.5429,
|
|
"mean_token_accuracy": 0.2094266712665558,
|
|
"num_tokens": 14644894.0,
|
|
"step": 6625
|
|
},
|
|
{
|
|
"entropy": 5.180797004699707,
|
|
"epoch": 6.683812405446293,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00018450000827851217,
|
|
"loss": 4.7109,
|
|
"mean_token_accuracy": 0.20146385878324508,
|
|
"num_tokens": 14658059.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 5.184048128128052,
|
|
"epoch": 6.688855269793242,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000184136984925952,
|
|
"loss": 4.495,
|
|
"mean_token_accuracy": 0.21745475381612778,
|
|
"num_tokens": 14668350.0,
|
|
"step": 6635
|
|
},
|
|
{
|
|
"entropy": 5.201734447479248,
|
|
"epoch": 6.693898134140191,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001837742439777386,
|
|
"loss": 4.5871,
|
|
"mean_token_accuracy": 0.20946993827819824,
|
|
"num_tokens": 14679700.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 5.157415819168091,
|
|
"epoch": 6.69894099848714,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00018341178656127948,
|
|
"loss": 4.5569,
|
|
"mean_token_accuracy": 0.211892569065094,
|
|
"num_tokens": 14690236.0,
|
|
"step": 6645
|
|
},
|
|
{
|
|
"entropy": 5.109434938430786,
|
|
"epoch": 6.7039838628340895,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00018304961380310038,
|
|
"loss": 4.5167,
|
|
"mean_token_accuracy": 0.21289398819208144,
|
|
"num_tokens": 14700361.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 5.211534595489502,
|
|
"epoch": 6.709026727181039,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00018268772682884273,
|
|
"loss": 4.6858,
|
|
"mean_token_accuracy": 0.20889659970998764,
|
|
"num_tokens": 14712262.0,
|
|
"step": 6655
|
|
},
|
|
{
|
|
"entropy": 5.203071165084839,
|
|
"epoch": 6.714069591527988,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00018232612676325944,
|
|
"loss": 4.6107,
|
|
"mean_token_accuracy": 0.20499747544527053,
|
|
"num_tokens": 14722242.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 5.18248438835144,
|
|
"epoch": 6.719112455874937,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001819648147302121,
|
|
"loss": 4.5723,
|
|
"mean_token_accuracy": 0.20836176872253417,
|
|
"num_tokens": 14733561.0,
|
|
"step": 6665
|
|
},
|
|
{
|
|
"entropy": 5.166212940216065,
|
|
"epoch": 6.724155320221886,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00018160379185266662,
|
|
"loss": 4.57,
|
|
"mean_token_accuracy": 0.20891642719507217,
|
|
"num_tokens": 14745227.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 5.171828985214233,
|
|
"epoch": 6.729198184568835,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00018124305925269058,
|
|
"loss": 4.6187,
|
|
"mean_token_accuracy": 0.2036039188504219,
|
|
"num_tokens": 14756812.0,
|
|
"step": 6675
|
|
},
|
|
{
|
|
"entropy": 5.174170398712159,
|
|
"epoch": 6.734241048915784,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00018088261805144918,
|
|
"loss": 4.6133,
|
|
"mean_token_accuracy": 0.2072407931089401,
|
|
"num_tokens": 14767920.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 5.218884086608886,
|
|
"epoch": 6.739283913262733,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00018052246936920192,
|
|
"loss": 4.5977,
|
|
"mean_token_accuracy": 0.20663738250732422,
|
|
"num_tokens": 14780314.0,
|
|
"step": 6685
|
|
},
|
|
{
|
|
"entropy": 5.144815444946289,
|
|
"epoch": 6.7443267776096825,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00018016261432529928,
|
|
"loss": 4.5852,
|
|
"mean_token_accuracy": 0.21311117857694625,
|
|
"num_tokens": 14790757.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 5.1522979736328125,
|
|
"epoch": 6.749369641956632,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00017980305403817893,
|
|
"loss": 4.5427,
|
|
"mean_token_accuracy": 0.21685515642166137,
|
|
"num_tokens": 14801152.0,
|
|
"step": 6695
|
|
},
|
|
{
|
|
"entropy": 5.121072816848755,
|
|
"epoch": 6.754412506303581,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00017944378962536279,
|
|
"loss": 4.5324,
|
|
"mean_token_accuracy": 0.2169053941965103,
|
|
"num_tokens": 14812422.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 5.141156387329102,
|
|
"epoch": 6.75945537065053,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00017908482220345268,
|
|
"loss": 4.5223,
|
|
"mean_token_accuracy": 0.2117744728922844,
|
|
"num_tokens": 14823062.0,
|
|
"step": 6705
|
|
},
|
|
{
|
|
"entropy": 5.141294956207275,
|
|
"epoch": 6.764498234997479,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00017872615288812787,
|
|
"loss": 4.5801,
|
|
"mean_token_accuracy": 0.21257184743881224,
|
|
"num_tokens": 14834247.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 5.14876356124878,
|
|
"epoch": 6.769541099344428,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00017836778279414057,
|
|
"loss": 4.5863,
|
|
"mean_token_accuracy": 0.20656152367591857,
|
|
"num_tokens": 14845538.0,
|
|
"step": 6715
|
|
},
|
|
{
|
|
"entropy": 5.208459520339966,
|
|
"epoch": 6.774583963691377,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00017800971303531354,
|
|
"loss": 4.6593,
|
|
"mean_token_accuracy": 0.19993465542793273,
|
|
"num_tokens": 14857597.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 5.1490837097167965,
|
|
"epoch": 6.779626828038325,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001776519447245356,
|
|
"loss": 4.6192,
|
|
"mean_token_accuracy": 0.2042725622653961,
|
|
"num_tokens": 14868599.0,
|
|
"step": 6725
|
|
},
|
|
{
|
|
"entropy": 5.197770023345948,
|
|
"epoch": 6.7846696923852745,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00017729447897375927,
|
|
"loss": 4.5701,
|
|
"mean_token_accuracy": 0.21612185835838318,
|
|
"num_tokens": 14879257.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 5.219246292114258,
|
|
"epoch": 6.789712556732224,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00017693731689399605,
|
|
"loss": 4.5824,
|
|
"mean_token_accuracy": 0.21452093720436097,
|
|
"num_tokens": 14890549.0,
|
|
"step": 6735
|
|
},
|
|
{
|
|
"entropy": 5.1342981338500975,
|
|
"epoch": 6.794755421079173,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00017658045959531416,
|
|
"loss": 4.5827,
|
|
"mean_token_accuracy": 0.2113274082541466,
|
|
"num_tokens": 14900946.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 5.18452525138855,
|
|
"epoch": 6.799798285426122,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00017622390818683404,
|
|
"loss": 4.5638,
|
|
"mean_token_accuracy": 0.21580399870872496,
|
|
"num_tokens": 14911430.0,
|
|
"step": 6745
|
|
},
|
|
{
|
|
"entropy": 5.1714800834655765,
|
|
"epoch": 6.804841149773071,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00017586766377672593,
|
|
"loss": 4.5744,
|
|
"mean_token_accuracy": 0.21286627501249314,
|
|
"num_tokens": 14921110.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 5.218984460830688,
|
|
"epoch": 6.80988401412002,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00017551172747220552,
|
|
"loss": 4.6276,
|
|
"mean_token_accuracy": 0.20382657945156096,
|
|
"num_tokens": 14931249.0,
|
|
"step": 6755
|
|
},
|
|
{
|
|
"entropy": 5.219883632659912,
|
|
"epoch": 6.814926878466969,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00017515610037953112,
|
|
"loss": 4.6138,
|
|
"mean_token_accuracy": 0.20429412126541138,
|
|
"num_tokens": 14942914.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 5.175330066680909,
|
|
"epoch": 6.819969742813918,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001748007836039999,
|
|
"loss": 4.5559,
|
|
"mean_token_accuracy": 0.2148611515760422,
|
|
"num_tokens": 14952863.0,
|
|
"step": 6765
|
|
},
|
|
{
|
|
"entropy": 5.155924987792969,
|
|
"epoch": 6.825012607160867,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00017444577824994475,
|
|
"loss": 4.6332,
|
|
"mean_token_accuracy": 0.20793980211019517,
|
|
"num_tokens": 14964027.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 5.212174654006958,
|
|
"epoch": 6.8300554715078166,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001740910854207303,
|
|
"loss": 4.5892,
|
|
"mean_token_accuracy": 0.2093237429857254,
|
|
"num_tokens": 14975139.0,
|
|
"step": 6775
|
|
},
|
|
{
|
|
"entropy": 5.125215816497803,
|
|
"epoch": 6.835098335854766,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00017373670621875016,
|
|
"loss": 4.5575,
|
|
"mean_token_accuracy": 0.21579992026090622,
|
|
"num_tokens": 14986481.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 5.2577534198760985,
|
|
"epoch": 6.840141200201715,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000173382641745423,
|
|
"loss": 4.5873,
|
|
"mean_token_accuracy": 0.20318973064422607,
|
|
"num_tokens": 14997897.0,
|
|
"step": 6785
|
|
},
|
|
{
|
|
"entropy": 5.194232559204101,
|
|
"epoch": 6.845184064548664,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001730288931011893,
|
|
"loss": 4.5553,
|
|
"mean_token_accuracy": 0.20918940752744675,
|
|
"num_tokens": 15007352.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 5.10802526473999,
|
|
"epoch": 6.850226928895613,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00017267546138550823,
|
|
"loss": 4.614,
|
|
"mean_token_accuracy": 0.20677233934402467,
|
|
"num_tokens": 15018489.0,
|
|
"step": 6795
|
|
},
|
|
{
|
|
"entropy": 5.1211706638336185,
|
|
"epoch": 6.855269793242562,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00017232234769685358,
|
|
"loss": 4.5609,
|
|
"mean_token_accuracy": 0.20823736488819122,
|
|
"num_tokens": 15029221.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 5.202625799179077,
|
|
"epoch": 6.86031265758951,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00017196955313271086,
|
|
"loss": 4.6403,
|
|
"mean_token_accuracy": 0.20660948008298874,
|
|
"num_tokens": 15039583.0,
|
|
"step": 6805
|
|
},
|
|
{
|
|
"entropy": 5.170046854019165,
|
|
"epoch": 6.8653555219364595,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00017161707878957368,
|
|
"loss": 4.5922,
|
|
"mean_token_accuracy": 0.21040977239608766,
|
|
"num_tokens": 15050694.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 5.164877986907959,
|
|
"epoch": 6.870398386283409,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00017126492576294053,
|
|
"loss": 4.5773,
|
|
"mean_token_accuracy": 0.20495762825012206,
|
|
"num_tokens": 15061155.0,
|
|
"step": 6815
|
|
},
|
|
{
|
|
"entropy": 5.177333688735962,
|
|
"epoch": 6.875441250630358,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001709130951473109,
|
|
"loss": 4.5632,
|
|
"mean_token_accuracy": 0.20725746005773543,
|
|
"num_tokens": 15070995.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 5.129806709289551,
|
|
"epoch": 6.880484114977307,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00017056158803618277,
|
|
"loss": 4.5625,
|
|
"mean_token_accuracy": 0.20937867909669877,
|
|
"num_tokens": 15081267.0,
|
|
"step": 6825
|
|
},
|
|
{
|
|
"entropy": 5.139181804656983,
|
|
"epoch": 6.885526979324256,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00017021040552204814,
|
|
"loss": 4.5642,
|
|
"mean_token_accuracy": 0.20684209167957307,
|
|
"num_tokens": 15091632.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 5.249155330657959,
|
|
"epoch": 6.890569843671205,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00016985954869639048,
|
|
"loss": 4.7049,
|
|
"mean_token_accuracy": 0.20422360002994538,
|
|
"num_tokens": 15103336.0,
|
|
"step": 6835
|
|
},
|
|
{
|
|
"entropy": 5.1753401279449465,
|
|
"epoch": 6.895612708018154,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00016950901864968076,
|
|
"loss": 4.5666,
|
|
"mean_token_accuracy": 0.21196467131376268,
|
|
"num_tokens": 15115109.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 5.1311005592346195,
|
|
"epoch": 6.900655572365103,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00016915881647137465,
|
|
"loss": 4.4859,
|
|
"mean_token_accuracy": 0.22543624937534332,
|
|
"num_tokens": 15125537.0,
|
|
"step": 6845
|
|
},
|
|
{
|
|
"entropy": 5.106833744049072,
|
|
"epoch": 6.905698436712052,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001688089432499084,
|
|
"loss": 4.5923,
|
|
"mean_token_accuracy": 0.20669917911291122,
|
|
"num_tokens": 15135714.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 5.125957393646241,
|
|
"epoch": 6.9107413010590015,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00016845940007269617,
|
|
"loss": 4.5335,
|
|
"mean_token_accuracy": 0.21664437353610994,
|
|
"num_tokens": 15145239.0,
|
|
"step": 6855
|
|
},
|
|
{
|
|
"entropy": 5.137819242477417,
|
|
"epoch": 6.915784165405951,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00016811018802612615,
|
|
"loss": 4.6421,
|
|
"mean_token_accuracy": 0.20316942781209946,
|
|
"num_tokens": 15156833.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 5.151161098480225,
|
|
"epoch": 6.9208270297529,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00016776130819555758,
|
|
"loss": 4.571,
|
|
"mean_token_accuracy": 0.2106426402926445,
|
|
"num_tokens": 15167606.0,
|
|
"step": 6865
|
|
},
|
|
{
|
|
"entropy": 5.1341265678405765,
|
|
"epoch": 6.925869894099849,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00016741276166531684,
|
|
"loss": 4.59,
|
|
"mean_token_accuracy": 0.20771575421094896,
|
|
"num_tokens": 15179086.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 5.128192138671875,
|
|
"epoch": 6.930912758446798,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00016706454951869468,
|
|
"loss": 4.5577,
|
|
"mean_token_accuracy": 0.21076772660017012,
|
|
"num_tokens": 15189566.0,
|
|
"step": 6875
|
|
},
|
|
{
|
|
"entropy": 5.163235187530518,
|
|
"epoch": 6.935955622793747,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00016671667283794245,
|
|
"loss": 4.5471,
|
|
"mean_token_accuracy": 0.2130429580807686,
|
|
"num_tokens": 15201108.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 5.199871587753296,
|
|
"epoch": 6.940998487140696,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00016636913270426888,
|
|
"loss": 4.6558,
|
|
"mean_token_accuracy": 0.20184287130832673,
|
|
"num_tokens": 15211894.0,
|
|
"step": 6885
|
|
},
|
|
{
|
|
"entropy": 5.173501634597779,
|
|
"epoch": 6.946041351487645,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00016602193019783673,
|
|
"loss": 4.557,
|
|
"mean_token_accuracy": 0.20634147077798842,
|
|
"num_tokens": 15222062.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 5.206723260879516,
|
|
"epoch": 6.9510842158345945,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00016567506639775947,
|
|
"loss": 4.6628,
|
|
"mean_token_accuracy": 0.19872487783432008,
|
|
"num_tokens": 15233633.0,
|
|
"step": 6895
|
|
},
|
|
{
|
|
"entropy": 5.157608985900879,
|
|
"epoch": 6.956127080181544,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00016532854238209783,
|
|
"loss": 4.6261,
|
|
"mean_token_accuracy": 0.19951253682374953,
|
|
"num_tokens": 15244831.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 5.190638160705566,
|
|
"epoch": 6.961169944528492,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001649823592278563,
|
|
"loss": 4.6489,
|
|
"mean_token_accuracy": 0.19773684144020082,
|
|
"num_tokens": 15255720.0,
|
|
"step": 6905
|
|
},
|
|
{
|
|
"entropy": 5.194908809661865,
|
|
"epoch": 6.966212808875441,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00016463651801098023,
|
|
"loss": 4.6186,
|
|
"mean_token_accuracy": 0.2155260130763054,
|
|
"num_tokens": 15267637.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 5.1314740657806395,
|
|
"epoch": 6.97125567322239,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00016429101980635193,
|
|
"loss": 4.5234,
|
|
"mean_token_accuracy": 0.20931103825569153,
|
|
"num_tokens": 15277604.0,
|
|
"step": 6915
|
|
},
|
|
{
|
|
"entropy": 5.152788972854614,
|
|
"epoch": 6.976298537569339,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00016394586568778795,
|
|
"loss": 4.5754,
|
|
"mean_token_accuracy": 0.21197256296873093,
|
|
"num_tokens": 15287922.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 5.120962524414063,
|
|
"epoch": 6.981341401916288,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00016360105672803515,
|
|
"loss": 4.5276,
|
|
"mean_token_accuracy": 0.2185438707470894,
|
|
"num_tokens": 15299539.0,
|
|
"step": 6925
|
|
},
|
|
{
|
|
"entropy": 5.229609203338623,
|
|
"epoch": 6.986384266263237,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00016325659399876792,
|
|
"loss": 4.6062,
|
|
"mean_token_accuracy": 0.19885406494140626,
|
|
"num_tokens": 15311023.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 5.212070560455322,
|
|
"epoch": 6.9914271306101865,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00016291247857058412,
|
|
"loss": 4.6358,
|
|
"mean_token_accuracy": 0.2048345386981964,
|
|
"num_tokens": 15323645.0,
|
|
"step": 6935
|
|
},
|
|
{
|
|
"entropy": 5.1193641185760494,
|
|
"epoch": 6.996469994957136,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00016256871151300268,
|
|
"loss": 4.5408,
|
|
"mean_token_accuracy": 0.2132187157869339,
|
|
"num_tokens": 15333887.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 5.2199704382154675,
|
|
"epoch": 7.00100857286939,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00016222529389445928,
|
|
"loss": 4.7588,
|
|
"mean_token_accuracy": 0.20126625895500183,
|
|
"num_tokens": 15344034.0,
|
|
"step": 6945
|
|
},
|
|
{
|
|
"entropy": 5.12688455581665,
|
|
"epoch": 7.006051437216339,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00016188222678230395,
|
|
"loss": 4.4199,
|
|
"mean_token_accuracy": 0.22685324251651764,
|
|
"num_tokens": 15353552.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 5.144989919662476,
|
|
"epoch": 7.011094301563288,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001615395112427972,
|
|
"loss": 4.4131,
|
|
"mean_token_accuracy": 0.22271778136491777,
|
|
"num_tokens": 15364332.0,
|
|
"step": 6955
|
|
},
|
|
{
|
|
"entropy": 5.1694964408874515,
|
|
"epoch": 7.016137165910237,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001611971483411069,
|
|
"loss": 4.4387,
|
|
"mean_token_accuracy": 0.2219022884964943,
|
|
"num_tokens": 15373990.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 5.170229911804199,
|
|
"epoch": 7.021180030257186,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00016085513914130463,
|
|
"loss": 4.4694,
|
|
"mean_token_accuracy": 0.2241543784737587,
|
|
"num_tokens": 15385088.0,
|
|
"step": 6965
|
|
},
|
|
{
|
|
"entropy": 5.141489696502686,
|
|
"epoch": 7.026222894604135,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00016051348470636306,
|
|
"loss": 4.4436,
|
|
"mean_token_accuracy": 0.22193147838115693,
|
|
"num_tokens": 15396210.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 5.164767408370972,
|
|
"epoch": 7.0312657589510845,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00016017218609815195,
|
|
"loss": 4.4355,
|
|
"mean_token_accuracy": 0.22168179452419282,
|
|
"num_tokens": 15406448.0,
|
|
"step": 6975
|
|
},
|
|
{
|
|
"entropy": 5.125365304946899,
|
|
"epoch": 7.036308623298034,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001598312443774353,
|
|
"loss": 4.4421,
|
|
"mean_token_accuracy": 0.21773712038993837,
|
|
"num_tokens": 15416570.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 5.1646605968475345,
|
|
"epoch": 7.041351487644983,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00015949066060386784,
|
|
"loss": 4.4626,
|
|
"mean_token_accuracy": 0.2192896842956543,
|
|
"num_tokens": 15426815.0,
|
|
"step": 6985
|
|
},
|
|
{
|
|
"entropy": 5.158516979217529,
|
|
"epoch": 7.046394351991931,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00015915043583599183,
|
|
"loss": 4.4568,
|
|
"mean_token_accuracy": 0.22245998680591583,
|
|
"num_tokens": 15437579.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 5.160056018829346,
|
|
"epoch": 7.05143721633888,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001588105711312336,
|
|
"loss": 4.4915,
|
|
"mean_token_accuracy": 0.2160475194454193,
|
|
"num_tokens": 15448194.0,
|
|
"step": 6995
|
|
},
|
|
{
|
|
"entropy": 5.15237283706665,
|
|
"epoch": 7.056480080685829,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00015847106754590055,
|
|
"loss": 4.429,
|
|
"mean_token_accuracy": 0.22077992260456086,
|
|
"num_tokens": 15458626.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"epoch": 7.056480080685829,
|
|
"eval_entropy": 5.115679789438806,
|
|
"eval_loss": 5.701469898223877,
|
|
"eval_mean_token_accuracy": 0.1547819839709891,
|
|
"eval_num_tokens": 15458626.0,
|
|
"eval_runtime": 2.6859,
|
|
"eval_samples_per_second": 1168.304,
|
|
"eval_steps_per_second": 146.317,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"entropy": 5.111939811706543,
|
|
"epoch": 7.061522945032778,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001581319261351777,
|
|
"loss": 4.431,
|
|
"mean_token_accuracy": 0.22781239449977875,
|
|
"num_tokens": 15469866.0,
|
|
"step": 7005
|
|
},
|
|
{
|
|
"entropy": 5.168635129928589,
|
|
"epoch": 7.0665658093797274,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00015779314795312426,
|
|
"loss": 4.489,
|
|
"mean_token_accuracy": 0.21452077925205232,
|
|
"num_tokens": 15481497.0,
|
|
"step": 7010
|
|
},
|
|
{
|
|
"entropy": 5.127896308898926,
|
|
"epoch": 7.071608673726677,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001574547340526707,
|
|
"loss": 4.416,
|
|
"mean_token_accuracy": 0.2185455396771431,
|
|
"num_tokens": 15493170.0,
|
|
"step": 7015
|
|
},
|
|
{
|
|
"entropy": 5.127766752243042,
|
|
"epoch": 7.076651538073626,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00015711668548561527,
|
|
"loss": 4.493,
|
|
"mean_token_accuracy": 0.2185678482055664,
|
|
"num_tokens": 15504122.0,
|
|
"step": 7020
|
|
},
|
|
{
|
|
"entropy": 5.105100393295288,
|
|
"epoch": 7.081694402420575,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00015677900330262075,
|
|
"loss": 4.3773,
|
|
"mean_token_accuracy": 0.22492315918207167,
|
|
"num_tokens": 15516568.0,
|
|
"step": 7025
|
|
},
|
|
{
|
|
"entropy": 5.200638151168823,
|
|
"epoch": 7.086737266767524,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00015644168855321098,
|
|
"loss": 4.4784,
|
|
"mean_token_accuracy": 0.20952097326517105,
|
|
"num_tokens": 15527386.0,
|
|
"step": 7030
|
|
},
|
|
{
|
|
"entropy": 5.167358016967773,
|
|
"epoch": 7.091780131114473,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00015610474228576818,
|
|
"loss": 4.4258,
|
|
"mean_token_accuracy": 0.2241312488913536,
|
|
"num_tokens": 15537738.0,
|
|
"step": 7035
|
|
},
|
|
{
|
|
"entropy": 5.135108566284179,
|
|
"epoch": 7.096822995461422,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00015576816554752892,
|
|
"loss": 4.4622,
|
|
"mean_token_accuracy": 0.22096373289823532,
|
|
"num_tokens": 15548257.0,
|
|
"step": 7040
|
|
},
|
|
{
|
|
"entropy": 5.125024223327637,
|
|
"epoch": 7.101865859808371,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001554319593845816,
|
|
"loss": 4.4222,
|
|
"mean_token_accuracy": 0.22111179828643798,
|
|
"num_tokens": 15558164.0,
|
|
"step": 7045
|
|
},
|
|
{
|
|
"entropy": 5.19545464515686,
|
|
"epoch": 7.10690872415532,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001550961248418627,
|
|
"loss": 4.4666,
|
|
"mean_token_accuracy": 0.220587882399559,
|
|
"num_tokens": 15569546.0,
|
|
"step": 7050
|
|
},
|
|
{
|
|
"entropy": 5.174524927139283,
|
|
"epoch": 7.1119515885022695,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00015476066296315376,
|
|
"loss": 4.4452,
|
|
"mean_token_accuracy": 0.2193372517824173,
|
|
"num_tokens": 15580125.0,
|
|
"step": 7055
|
|
},
|
|
{
|
|
"entropy": 5.1802925109863285,
|
|
"epoch": 7.116994452849219,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000154425574791078,
|
|
"loss": 4.4091,
|
|
"mean_token_accuracy": 0.23008309453725814,
|
|
"num_tokens": 15589978.0,
|
|
"step": 7060
|
|
},
|
|
{
|
|
"entropy": 5.126687812805176,
|
|
"epoch": 7.122037317196168,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00015409086136709723,
|
|
"loss": 4.3996,
|
|
"mean_token_accuracy": 0.22829253524541854,
|
|
"num_tokens": 15600114.0,
|
|
"step": 7065
|
|
},
|
|
{
|
|
"entropy": 5.1117795467376705,
|
|
"epoch": 7.127080181543117,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001537565237315084,
|
|
"loss": 4.4203,
|
|
"mean_token_accuracy": 0.21933439373970032,
|
|
"num_tokens": 15611976.0,
|
|
"step": 7070
|
|
},
|
|
{
|
|
"entropy": 5.172864770889282,
|
|
"epoch": 7.132123045890066,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00015342256292344065,
|
|
"loss": 4.4018,
|
|
"mean_token_accuracy": 0.2312544107437134,
|
|
"num_tokens": 15622036.0,
|
|
"step": 7075
|
|
},
|
|
{
|
|
"entropy": 5.1391348361969,
|
|
"epoch": 7.137165910237014,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00015308897998085187,
|
|
"loss": 4.4822,
|
|
"mean_token_accuracy": 0.21364797204732894,
|
|
"num_tokens": 15633874.0,
|
|
"step": 7080
|
|
},
|
|
{
|
|
"entropy": 5.2143370628356935,
|
|
"epoch": 7.142208774583963,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00015275577594052564,
|
|
"loss": 4.4576,
|
|
"mean_token_accuracy": 0.22128628939390182,
|
|
"num_tokens": 15643580.0,
|
|
"step": 7085
|
|
},
|
|
{
|
|
"entropy": 5.195936679840088,
|
|
"epoch": 7.147251638930912,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001524229518380676,
|
|
"loss": 4.4631,
|
|
"mean_token_accuracy": 0.2238822802901268,
|
|
"num_tokens": 15654672.0,
|
|
"step": 7090
|
|
},
|
|
{
|
|
"entropy": 5.197504234313965,
|
|
"epoch": 7.1522945032778615,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001520905087079029,
|
|
"loss": 4.532,
|
|
"mean_token_accuracy": 0.21360340416431428,
|
|
"num_tokens": 15666672.0,
|
|
"step": 7095
|
|
},
|
|
{
|
|
"entropy": 5.190901088714599,
|
|
"epoch": 7.157337367624811,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00015175844758327239,
|
|
"loss": 4.472,
|
|
"mean_token_accuracy": 0.22510324269533158,
|
|
"num_tokens": 15676544.0,
|
|
"step": 7100
|
|
},
|
|
{
|
|
"entropy": 5.1282518863677975,
|
|
"epoch": 7.16238023197176,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00015142676949622973,
|
|
"loss": 4.4422,
|
|
"mean_token_accuracy": 0.21841446608304976,
|
|
"num_tokens": 15686798.0,
|
|
"step": 7105
|
|
},
|
|
{
|
|
"entropy": 5.173515796661377,
|
|
"epoch": 7.167423096318709,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00015109547547763802,
|
|
"loss": 4.4284,
|
|
"mean_token_accuracy": 0.21863365322351455,
|
|
"num_tokens": 15696521.0,
|
|
"step": 7110
|
|
},
|
|
{
|
|
"entropy": 5.170062160491943,
|
|
"epoch": 7.172465960665658,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00015076456655716675,
|
|
"loss": 4.4606,
|
|
"mean_token_accuracy": 0.2137729749083519,
|
|
"num_tokens": 15708445.0,
|
|
"step": 7115
|
|
},
|
|
{
|
|
"entropy": 5.154700708389282,
|
|
"epoch": 7.177508825012607,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00015043404376328856,
|
|
"loss": 4.4345,
|
|
"mean_token_accuracy": 0.2191716343164444,
|
|
"num_tokens": 15720584.0,
|
|
"step": 7120
|
|
},
|
|
{
|
|
"entropy": 5.173767852783203,
|
|
"epoch": 7.182551689359556,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00015010390812327578,
|
|
"loss": 4.5326,
|
|
"mean_token_accuracy": 0.21602841466665268,
|
|
"num_tokens": 15732113.0,
|
|
"step": 7125
|
|
},
|
|
{
|
|
"entropy": 5.150187635421753,
|
|
"epoch": 7.187594553706505,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001497741606631977,
|
|
"loss": 4.4526,
|
|
"mean_token_accuracy": 0.22465599924325944,
|
|
"num_tokens": 15744217.0,
|
|
"step": 7130
|
|
},
|
|
{
|
|
"entropy": 5.202928256988526,
|
|
"epoch": 7.1926374180534545,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001494448024079169,
|
|
"loss": 4.4843,
|
|
"mean_token_accuracy": 0.21621037125587464,
|
|
"num_tokens": 15754970.0,
|
|
"step": 7135
|
|
},
|
|
{
|
|
"entropy": 5.1663435935974125,
|
|
"epoch": 7.197680282400404,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00014911583438108656,
|
|
"loss": 4.503,
|
|
"mean_token_accuracy": 0.21592814326286316,
|
|
"num_tokens": 15765117.0,
|
|
"step": 7140
|
|
},
|
|
{
|
|
"entropy": 5.143350505828858,
|
|
"epoch": 7.202723146747353,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00014878725760514682,
|
|
"loss": 4.4493,
|
|
"mean_token_accuracy": 0.2220911517739296,
|
|
"num_tokens": 15775685.0,
|
|
"step": 7145
|
|
},
|
|
{
|
|
"entropy": 5.217806959152222,
|
|
"epoch": 7.207766011094302,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00014845907310132202,
|
|
"loss": 4.487,
|
|
"mean_token_accuracy": 0.21625738143920897,
|
|
"num_tokens": 15785392.0,
|
|
"step": 7150
|
|
},
|
|
{
|
|
"entropy": 5.154491567611695,
|
|
"epoch": 7.212808875441251,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00014813128188961695,
|
|
"loss": 4.4868,
|
|
"mean_token_accuracy": 0.2222212553024292,
|
|
"num_tokens": 15798216.0,
|
|
"step": 7155
|
|
},
|
|
{
|
|
"entropy": 5.117153024673462,
|
|
"epoch": 7.2178517397882,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00014780388498881447,
|
|
"loss": 4.4446,
|
|
"mean_token_accuracy": 0.21864890158176423,
|
|
"num_tokens": 15808457.0,
|
|
"step": 7160
|
|
},
|
|
{
|
|
"entropy": 5.123981618881226,
|
|
"epoch": 7.222894604135149,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001474768834164715,
|
|
"loss": 4.458,
|
|
"mean_token_accuracy": 0.22008757591247557,
|
|
"num_tokens": 15820572.0,
|
|
"step": 7165
|
|
},
|
|
{
|
|
"entropy": 5.217930507659912,
|
|
"epoch": 7.227937468482098,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00014715027818891656,
|
|
"loss": 4.4602,
|
|
"mean_token_accuracy": 0.21566811800003052,
|
|
"num_tokens": 15831969.0,
|
|
"step": 7170
|
|
},
|
|
{
|
|
"entropy": 5.137494277954102,
|
|
"epoch": 7.2329803328290465,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001468240703212463,
|
|
"loss": 4.504,
|
|
"mean_token_accuracy": 0.22035356163978576,
|
|
"num_tokens": 15843973.0,
|
|
"step": 7175
|
|
},
|
|
{
|
|
"entropy": 5.236536073684692,
|
|
"epoch": 7.238023197175996,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00014649826082732226,
|
|
"loss": 4.5679,
|
|
"mean_token_accuracy": 0.21232356876134872,
|
|
"num_tokens": 15855908.0,
|
|
"step": 7180
|
|
},
|
|
{
|
|
"entropy": 5.140198135375977,
|
|
"epoch": 7.243066061522945,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00014617285071976777,
|
|
"loss": 4.4263,
|
|
"mean_token_accuracy": 0.22509616613388062,
|
|
"num_tokens": 15865653.0,
|
|
"step": 7185
|
|
},
|
|
{
|
|
"entropy": 5.136035776138305,
|
|
"epoch": 7.248108925869894,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00014584784100996512,
|
|
"loss": 4.5072,
|
|
"mean_token_accuracy": 0.22311645448207856,
|
|
"num_tokens": 15877109.0,
|
|
"step": 7190
|
|
},
|
|
{
|
|
"entropy": 5.169123744964599,
|
|
"epoch": 7.253151790216843,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001455232327080518,
|
|
"loss": 4.4964,
|
|
"mean_token_accuracy": 0.21839261054992676,
|
|
"num_tokens": 15888640.0,
|
|
"step": 7195
|
|
},
|
|
{
|
|
"entropy": 5.168622398376465,
|
|
"epoch": 7.258194654563792,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00014519902682291795,
|
|
"loss": 4.4887,
|
|
"mean_token_accuracy": 0.22024169862270354,
|
|
"num_tokens": 15900028.0,
|
|
"step": 7200
|
|
},
|
|
{
|
|
"entropy": 5.148393821716309,
|
|
"epoch": 7.263237518910741,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00014487522436220304,
|
|
"loss": 4.4751,
|
|
"mean_token_accuracy": 0.21593110561370848,
|
|
"num_tokens": 15911429.0,
|
|
"step": 7205
|
|
},
|
|
{
|
|
"entropy": 5.120872116088867,
|
|
"epoch": 7.26828038325769,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00014455182633229258,
|
|
"loss": 4.4383,
|
|
"mean_token_accuracy": 0.22649429738521576,
|
|
"num_tokens": 15923169.0,
|
|
"step": 7210
|
|
},
|
|
{
|
|
"entropy": 5.199803590774536,
|
|
"epoch": 7.273323247604639,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00014422883373831497,
|
|
"loss": 4.4978,
|
|
"mean_token_accuracy": 0.21979171633720399,
|
|
"num_tokens": 15935296.0,
|
|
"step": 7215
|
|
},
|
|
{
|
|
"entropy": 5.176680564880371,
|
|
"epoch": 7.278366111951589,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00014390624758413872,
|
|
"loss": 4.4587,
|
|
"mean_token_accuracy": 0.2210767835378647,
|
|
"num_tokens": 15946675.0,
|
|
"step": 7220
|
|
},
|
|
{
|
|
"entropy": 5.1491326808929445,
|
|
"epoch": 7.283408976298538,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00014358406887236914,
|
|
"loss": 4.4455,
|
|
"mean_token_accuracy": 0.21487119793891907,
|
|
"num_tokens": 15957645.0,
|
|
"step": 7225
|
|
},
|
|
{
|
|
"entropy": 5.149222421646118,
|
|
"epoch": 7.288451840645487,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00014326229860434488,
|
|
"loss": 4.4773,
|
|
"mean_token_accuracy": 0.2187603995203972,
|
|
"num_tokens": 15968958.0,
|
|
"step": 7230
|
|
},
|
|
{
|
|
"entropy": 5.139088773727417,
|
|
"epoch": 7.293494704992436,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001429409377801355,
|
|
"loss": 4.4251,
|
|
"mean_token_accuracy": 0.22115424871444703,
|
|
"num_tokens": 15979918.0,
|
|
"step": 7235
|
|
},
|
|
{
|
|
"entropy": 5.131740760803223,
|
|
"epoch": 7.298537569339385,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00014261998739853785,
|
|
"loss": 4.4739,
|
|
"mean_token_accuracy": 0.2160690024495125,
|
|
"num_tokens": 15991517.0,
|
|
"step": 7240
|
|
},
|
|
{
|
|
"entropy": 5.121307229995727,
|
|
"epoch": 7.303580433686334,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00014229944845707315,
|
|
"loss": 4.4052,
|
|
"mean_token_accuracy": 0.22437763810157776,
|
|
"num_tokens": 16003285.0,
|
|
"step": 7245
|
|
},
|
|
{
|
|
"entropy": 5.139844608306885,
|
|
"epoch": 7.308623298033283,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00014197932195198374,
|
|
"loss": 4.4278,
|
|
"mean_token_accuracy": 0.22730148434638978,
|
|
"num_tokens": 16014113.0,
|
|
"step": 7250
|
|
},
|
|
{
|
|
"entropy": 5.093283319473267,
|
|
"epoch": 7.313666162380232,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00014165960887823026,
|
|
"loss": 4.4768,
|
|
"mean_token_accuracy": 0.2244907259941101,
|
|
"num_tokens": 16025592.0,
|
|
"step": 7255
|
|
},
|
|
{
|
|
"entropy": 5.190410995483399,
|
|
"epoch": 7.318709026727181,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00014134031022948817,
|
|
"loss": 4.5311,
|
|
"mean_token_accuracy": 0.2123086705803871,
|
|
"num_tokens": 16037141.0,
|
|
"step": 7260
|
|
},
|
|
{
|
|
"entropy": 5.210147476196289,
|
|
"epoch": 7.32375189107413,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00014102142699814525,
|
|
"loss": 4.5889,
|
|
"mean_token_accuracy": 0.20612500607967377,
|
|
"num_tokens": 16047501.0,
|
|
"step": 7265
|
|
},
|
|
{
|
|
"entropy": 5.115164422988892,
|
|
"epoch": 7.328794755421079,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00014070296017529773,
|
|
"loss": 4.4074,
|
|
"mean_token_accuracy": 0.2320347845554352,
|
|
"num_tokens": 16058285.0,
|
|
"step": 7270
|
|
},
|
|
{
|
|
"entropy": 5.18541259765625,
|
|
"epoch": 7.333837619768028,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00014038491075074805,
|
|
"loss": 4.5291,
|
|
"mean_token_accuracy": 0.2199672818183899,
|
|
"num_tokens": 16070527.0,
|
|
"step": 7275
|
|
},
|
|
{
|
|
"entropy": 5.179767084121704,
|
|
"epoch": 7.338880484114977,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00014006727971300098,
|
|
"loss": 4.4985,
|
|
"mean_token_accuracy": 0.2213954895734787,
|
|
"num_tokens": 16081466.0,
|
|
"step": 7280
|
|
},
|
|
{
|
|
"entropy": 5.153761196136474,
|
|
"epoch": 7.343923348461926,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00013975006804926128,
|
|
"loss": 4.4317,
|
|
"mean_token_accuracy": 0.22199294567108155,
|
|
"num_tokens": 16091922.0,
|
|
"step": 7285
|
|
},
|
|
{
|
|
"entropy": 5.121073389053345,
|
|
"epoch": 7.348966212808875,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00013943327674542988,
|
|
"loss": 4.4304,
|
|
"mean_token_accuracy": 0.22336436212062835,
|
|
"num_tokens": 16103747.0,
|
|
"step": 7290
|
|
},
|
|
{
|
|
"entropy": 5.146017026901245,
|
|
"epoch": 7.354009077155824,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00013911690678610164,
|
|
"loss": 4.4309,
|
|
"mean_token_accuracy": 0.22008072584867477,
|
|
"num_tokens": 16113895.0,
|
|
"step": 7295
|
|
},
|
|
{
|
|
"entropy": 5.105138158798217,
|
|
"epoch": 7.3590519415027735,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00013880095915456167,
|
|
"loss": 4.4131,
|
|
"mean_token_accuracy": 0.22727687656879425,
|
|
"num_tokens": 16122954.0,
|
|
"step": 7300
|
|
},
|
|
{
|
|
"entropy": 5.116324329376221,
|
|
"epoch": 7.364094805849723,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00013848543483278256,
|
|
"loss": 4.4335,
|
|
"mean_token_accuracy": 0.22294033020734788,
|
|
"num_tokens": 16133932.0,
|
|
"step": 7305
|
|
},
|
|
{
|
|
"entropy": 5.191923332214356,
|
|
"epoch": 7.369137670196672,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000138170334801421,
|
|
"loss": 4.5362,
|
|
"mean_token_accuracy": 0.2144575223326683,
|
|
"num_tokens": 16144491.0,
|
|
"step": 7310
|
|
},
|
|
{
|
|
"entropy": 5.123883533477783,
|
|
"epoch": 7.374180534543621,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00013785566003981538,
|
|
"loss": 4.414,
|
|
"mean_token_accuracy": 0.2232085272669792,
|
|
"num_tokens": 16155328.0,
|
|
"step": 7315
|
|
},
|
|
{
|
|
"entropy": 5.120877313613891,
|
|
"epoch": 7.37922339889057,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001375414115259819,
|
|
"loss": 4.5163,
|
|
"mean_token_accuracy": 0.21946289539337158,
|
|
"num_tokens": 16166642.0,
|
|
"step": 7320
|
|
},
|
|
{
|
|
"entropy": 5.221355772018432,
|
|
"epoch": 7.384266263237519,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001372275902366122,
|
|
"loss": 4.4633,
|
|
"mean_token_accuracy": 0.21824126392602922,
|
|
"num_tokens": 16177736.0,
|
|
"step": 7325
|
|
},
|
|
{
|
|
"entropy": 5.146222400665283,
|
|
"epoch": 7.389309127584468,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00013691419714707026,
|
|
"loss": 4.5023,
|
|
"mean_token_accuracy": 0.21474253833293916,
|
|
"num_tokens": 16188342.0,
|
|
"step": 7330
|
|
},
|
|
{
|
|
"entropy": 5.128671073913575,
|
|
"epoch": 7.394351991931417,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00013660123323138893,
|
|
"loss": 4.4457,
|
|
"mean_token_accuracy": 0.2214590921998024,
|
|
"num_tokens": 16200259.0,
|
|
"step": 7335
|
|
},
|
|
{
|
|
"entropy": 5.197023868560791,
|
|
"epoch": 7.3993948562783665,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00013628869946226738,
|
|
"loss": 4.5164,
|
|
"mean_token_accuracy": 0.21630325466394423,
|
|
"num_tokens": 16214131.0,
|
|
"step": 7340
|
|
},
|
|
{
|
|
"entropy": 5.169191551208496,
|
|
"epoch": 7.404437720625316,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001359765968110676,
|
|
"loss": 4.4901,
|
|
"mean_token_accuracy": 0.2153078496456146,
|
|
"num_tokens": 16224382.0,
|
|
"step": 7345
|
|
},
|
|
{
|
|
"entropy": 5.17403826713562,
|
|
"epoch": 7.409480584972265,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00013566492624781198,
|
|
"loss": 4.4903,
|
|
"mean_token_accuracy": 0.21905571818351746,
|
|
"num_tokens": 16237212.0,
|
|
"step": 7350
|
|
},
|
|
{
|
|
"entropy": 5.188941955566406,
|
|
"epoch": 7.414523449319213,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001353536887411797,
|
|
"loss": 4.5582,
|
|
"mean_token_accuracy": 0.2148658514022827,
|
|
"num_tokens": 16249487.0,
|
|
"step": 7355
|
|
},
|
|
{
|
|
"entropy": 5.155982780456543,
|
|
"epoch": 7.419566313666162,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00013504288525850416,
|
|
"loss": 4.4054,
|
|
"mean_token_accuracy": 0.22213341891765595,
|
|
"num_tokens": 16259782.0,
|
|
"step": 7360
|
|
},
|
|
{
|
|
"entropy": 5.207163619995117,
|
|
"epoch": 7.424609178013111,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00013473251676576975,
|
|
"loss": 4.5571,
|
|
"mean_token_accuracy": 0.21379450410604478,
|
|
"num_tokens": 16271012.0,
|
|
"step": 7365
|
|
},
|
|
{
|
|
"entropy": 5.158382272720337,
|
|
"epoch": 7.42965204236006,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000134422584227609,
|
|
"loss": 4.5429,
|
|
"mean_token_accuracy": 0.21223240196704865,
|
|
"num_tokens": 16282282.0,
|
|
"step": 7370
|
|
},
|
|
{
|
|
"entropy": 5.17113561630249,
|
|
"epoch": 7.434694906707009,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001341130886072992,
|
|
"loss": 4.545,
|
|
"mean_token_accuracy": 0.2141255781054497,
|
|
"num_tokens": 16292966.0,
|
|
"step": 7375
|
|
},
|
|
{
|
|
"entropy": 5.099735116958618,
|
|
"epoch": 7.4397377710539585,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00013380403086676014,
|
|
"loss": 4.4284,
|
|
"mean_token_accuracy": 0.22266411930322647,
|
|
"num_tokens": 16304368.0,
|
|
"step": 7380
|
|
},
|
|
{
|
|
"entropy": 5.170955228805542,
|
|
"epoch": 7.444780635400908,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00013349541196655006,
|
|
"loss": 4.46,
|
|
"mean_token_accuracy": 0.2255719229578972,
|
|
"num_tokens": 16315916.0,
|
|
"step": 7385
|
|
},
|
|
{
|
|
"entropy": 5.137975692749023,
|
|
"epoch": 7.449823499747857,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00013318723286586388,
|
|
"loss": 4.4677,
|
|
"mean_token_accuracy": 0.22046116441488267,
|
|
"num_tokens": 16327642.0,
|
|
"step": 7390
|
|
},
|
|
{
|
|
"entropy": 5.141770076751709,
|
|
"epoch": 7.454866364094806,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001328794945225293,
|
|
"loss": 4.5001,
|
|
"mean_token_accuracy": 0.21883831173181534,
|
|
"num_tokens": 16339396.0,
|
|
"step": 7395
|
|
},
|
|
{
|
|
"entropy": 5.18817138671875,
|
|
"epoch": 7.459909228441755,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00013257219789300417,
|
|
"loss": 4.5122,
|
|
"mean_token_accuracy": 0.22510863691568375,
|
|
"num_tokens": 16350327.0,
|
|
"step": 7400
|
|
},
|
|
{
|
|
"entropy": 5.178559017181397,
|
|
"epoch": 7.464952092788704,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001322653439323735,
|
|
"loss": 4.5534,
|
|
"mean_token_accuracy": 0.20883676260709763,
|
|
"num_tokens": 16362102.0,
|
|
"step": 7405
|
|
},
|
|
{
|
|
"entropy": 5.188013076782227,
|
|
"epoch": 7.469994957135653,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00013195893359434658,
|
|
"loss": 4.5065,
|
|
"mean_token_accuracy": 0.21651567816734313,
|
|
"num_tokens": 16372244.0,
|
|
"step": 7410
|
|
},
|
|
{
|
|
"entropy": 5.174899768829346,
|
|
"epoch": 7.475037821482602,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00013165296783125357,
|
|
"loss": 4.5053,
|
|
"mean_token_accuracy": 0.21813054382801056,
|
|
"num_tokens": 16382814.0,
|
|
"step": 7415
|
|
},
|
|
{
|
|
"entropy": 5.072721242904663,
|
|
"epoch": 7.480080685829551,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00013134744759404332,
|
|
"loss": 4.4075,
|
|
"mean_token_accuracy": 0.227798530459404,
|
|
"num_tokens": 16394355.0,
|
|
"step": 7420
|
|
},
|
|
{
|
|
"entropy": 5.157034635543823,
|
|
"epoch": 7.4851235501765006,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001310423738322797,
|
|
"loss": 4.4515,
|
|
"mean_token_accuracy": 0.21789961010217668,
|
|
"num_tokens": 16406250.0,
|
|
"step": 7425
|
|
},
|
|
{
|
|
"entropy": 5.1012561321258545,
|
|
"epoch": 7.49016641452345,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00013073774749413903,
|
|
"loss": 4.409,
|
|
"mean_token_accuracy": 0.2250395029783249,
|
|
"num_tokens": 16417926.0,
|
|
"step": 7430
|
|
},
|
|
{
|
|
"entropy": 5.175812005996704,
|
|
"epoch": 7.495209278870399,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00013043356952640697,
|
|
"loss": 4.4837,
|
|
"mean_token_accuracy": 0.22154101133346557,
|
|
"num_tokens": 16428757.0,
|
|
"step": 7435
|
|
},
|
|
{
|
|
"entropy": 5.1692907333374025,
|
|
"epoch": 7.500252143217347,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00013012984087447567,
|
|
"loss": 4.4767,
|
|
"mean_token_accuracy": 0.22177896201610564,
|
|
"num_tokens": 16440955.0,
|
|
"step": 7440
|
|
},
|
|
{
|
|
"entropy": 5.189160871505737,
|
|
"epoch": 7.505295007564296,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00012982656248234088,
|
|
"loss": 4.5406,
|
|
"mean_token_accuracy": 0.20465246737003326,
|
|
"num_tokens": 16452448.0,
|
|
"step": 7445
|
|
},
|
|
{
|
|
"entropy": 5.141700267791748,
|
|
"epoch": 7.510337871911245,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00012952373529259863,
|
|
"loss": 4.4341,
|
|
"mean_token_accuracy": 0.22389615923166276,
|
|
"num_tokens": 16463219.0,
|
|
"step": 7450
|
|
},
|
|
{
|
|
"entropy": 5.177028846740723,
|
|
"epoch": 7.515380736258194,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00012922136024644298,
|
|
"loss": 4.4612,
|
|
"mean_token_accuracy": 0.22107804268598558,
|
|
"num_tokens": 16473295.0,
|
|
"step": 7455
|
|
},
|
|
{
|
|
"entropy": 5.189627265930175,
|
|
"epoch": 7.5204236006051435,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00012891943828366247,
|
|
"loss": 4.4551,
|
|
"mean_token_accuracy": 0.22378707230091094,
|
|
"num_tokens": 16484152.0,
|
|
"step": 7460
|
|
},
|
|
{
|
|
"entropy": 5.153527545928955,
|
|
"epoch": 7.525466464952093,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00012861797034263763,
|
|
"loss": 4.5485,
|
|
"mean_token_accuracy": 0.21869119107723237,
|
|
"num_tokens": 16495115.0,
|
|
"step": 7465
|
|
},
|
|
{
|
|
"entropy": 5.122735404968262,
|
|
"epoch": 7.530509329299042,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001283169573603376,
|
|
"loss": 4.4237,
|
|
"mean_token_accuracy": 0.22816676199436187,
|
|
"num_tokens": 16506032.0,
|
|
"step": 7470
|
|
},
|
|
{
|
|
"entropy": 5.175145387649536,
|
|
"epoch": 7.535552193645991,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00012801640027231792,
|
|
"loss": 4.4988,
|
|
"mean_token_accuracy": 0.21867337226867675,
|
|
"num_tokens": 16519268.0,
|
|
"step": 7475
|
|
},
|
|
{
|
|
"entropy": 5.171888923645019,
|
|
"epoch": 7.54059505799294,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00012771630001271668,
|
|
"loss": 4.5081,
|
|
"mean_token_accuracy": 0.21709107160568236,
|
|
"num_tokens": 16530166.0,
|
|
"step": 7480
|
|
},
|
|
{
|
|
"entropy": 5.156844234466552,
|
|
"epoch": 7.545637922339889,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00012741665751425253,
|
|
"loss": 4.4986,
|
|
"mean_token_accuracy": 0.21884409338235855,
|
|
"num_tokens": 16542010.0,
|
|
"step": 7485
|
|
},
|
|
{
|
|
"entropy": 5.190623188018799,
|
|
"epoch": 7.550680786686838,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00012711747370822127,
|
|
"loss": 4.4663,
|
|
"mean_token_accuracy": 0.22077708542346955,
|
|
"num_tokens": 16553523.0,
|
|
"step": 7490
|
|
},
|
|
{
|
|
"entropy": 5.190775966644287,
|
|
"epoch": 7.555723651033787,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00012681874952449315,
|
|
"loss": 4.5353,
|
|
"mean_token_accuracy": 0.2212969943881035,
|
|
"num_tokens": 16564149.0,
|
|
"step": 7495
|
|
},
|
|
{
|
|
"entropy": 5.176997566223145,
|
|
"epoch": 7.560766515380736,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001265204858915097,
|
|
"loss": 4.4509,
|
|
"mean_token_accuracy": 0.21607149988412858,
|
|
"num_tokens": 16574157.0,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"epoch": 7.560766515380736,
|
|
"eval_entropy": 5.0682855707998495,
|
|
"eval_loss": 5.700758457183838,
|
|
"eval_mean_token_accuracy": 0.15572236339933695,
|
|
"eval_num_tokens": 16574157.0,
|
|
"eval_runtime": 2.6486,
|
|
"eval_samples_per_second": 1184.786,
|
|
"eval_steps_per_second": 148.381,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"entropy": 5.147386503219605,
|
|
"epoch": 7.5658093797276855,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00012622268373628127,
|
|
"loss": 4.53,
|
|
"mean_token_accuracy": 0.21699445694684982,
|
|
"num_tokens": 16585571.0,
|
|
"step": 7505
|
|
},
|
|
{
|
|
"entropy": 5.11991286277771,
|
|
"epoch": 7.570852244074635,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00012592534398438369,
|
|
"loss": 4.3921,
|
|
"mean_token_accuracy": 0.2207753375172615,
|
|
"num_tokens": 16595790.0,
|
|
"step": 7510
|
|
},
|
|
{
|
|
"entropy": 5.137489891052246,
|
|
"epoch": 7.575895108421584,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00012562846755995586,
|
|
"loss": 4.4421,
|
|
"mean_token_accuracy": 0.21428816616535187,
|
|
"num_tokens": 16607529.0,
|
|
"step": 7515
|
|
},
|
|
{
|
|
"entropy": 5.1478346824646,
|
|
"epoch": 7.580937972768533,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00012533205538569654,
|
|
"loss": 4.4849,
|
|
"mean_token_accuracy": 0.22178708761930466,
|
|
"num_tokens": 16619162.0,
|
|
"step": 7520
|
|
},
|
|
{
|
|
"entropy": 5.084278917312622,
|
|
"epoch": 7.585980837115482,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001250361083828617,
|
|
"loss": 4.4353,
|
|
"mean_token_accuracy": 0.22726704329252242,
|
|
"num_tokens": 16629723.0,
|
|
"step": 7525
|
|
},
|
|
{
|
|
"entropy": 5.163690853118896,
|
|
"epoch": 7.591023701462431,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001247406274712613,
|
|
"loss": 4.5184,
|
|
"mean_token_accuracy": 0.21505824476480484,
|
|
"num_tokens": 16641480.0,
|
|
"step": 7530
|
|
},
|
|
{
|
|
"entropy": 5.2088484287261965,
|
|
"epoch": 7.59606656580938,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00012444561356925697,
|
|
"loss": 4.5196,
|
|
"mean_token_accuracy": 0.2133194088935852,
|
|
"num_tokens": 16651276.0,
|
|
"step": 7535
|
|
},
|
|
{
|
|
"entropy": 5.173453521728516,
|
|
"epoch": 7.6011094301563285,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00012415106759375855,
|
|
"loss": 4.4883,
|
|
"mean_token_accuracy": 0.21474240124225616,
|
|
"num_tokens": 16662245.0,
|
|
"step": 7540
|
|
},
|
|
{
|
|
"entropy": 5.172801542282104,
|
|
"epoch": 7.606152294503278,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00012385699046022188,
|
|
"loss": 4.5113,
|
|
"mean_token_accuracy": 0.21747588515281677,
|
|
"num_tokens": 16672206.0,
|
|
"step": 7545
|
|
},
|
|
{
|
|
"entropy": 5.151517677307129,
|
|
"epoch": 7.611195158850227,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00012356338308264537,
|
|
"loss": 4.4637,
|
|
"mean_token_accuracy": 0.22063014060258865,
|
|
"num_tokens": 16684734.0,
|
|
"step": 7550
|
|
},
|
|
{
|
|
"entropy": 5.096904373168945,
|
|
"epoch": 7.616238023197176,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001232702463735676,
|
|
"loss": 4.411,
|
|
"mean_token_accuracy": 0.2281380459666252,
|
|
"num_tokens": 16695801.0,
|
|
"step": 7555
|
|
},
|
|
{
|
|
"entropy": 5.133680248260498,
|
|
"epoch": 7.621280887544125,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00012297758124406427,
|
|
"loss": 4.4438,
|
|
"mean_token_accuracy": 0.22606228142976761,
|
|
"num_tokens": 16706629.0,
|
|
"step": 7560
|
|
},
|
|
{
|
|
"entropy": 5.2117695808410645,
|
|
"epoch": 7.626323751891074,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00012268538860374517,
|
|
"loss": 4.5389,
|
|
"mean_token_accuracy": 0.21765971034765244,
|
|
"num_tokens": 16718468.0,
|
|
"step": 7565
|
|
},
|
|
{
|
|
"entropy": 5.193145847320556,
|
|
"epoch": 7.631366616238023,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00012239366936075196,
|
|
"loss": 4.5135,
|
|
"mean_token_accuracy": 0.21309635788202286,
|
|
"num_tokens": 16729054.0,
|
|
"step": 7570
|
|
},
|
|
{
|
|
"entropy": 5.135074853897095,
|
|
"epoch": 7.636409480584972,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00012210242442175462,
|
|
"loss": 4.4465,
|
|
"mean_token_accuracy": 0.23070860505104065,
|
|
"num_tokens": 16739751.0,
|
|
"step": 7575
|
|
},
|
|
{
|
|
"entropy": 5.142823362350464,
|
|
"epoch": 7.641452344931921,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00012181165469194924,
|
|
"loss": 4.4904,
|
|
"mean_token_accuracy": 0.22042959779500962,
|
|
"num_tokens": 16751395.0,
|
|
"step": 7580
|
|
},
|
|
{
|
|
"entropy": 5.1433206558227536,
|
|
"epoch": 7.6464952092788705,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001215213610750548,
|
|
"loss": 4.4969,
|
|
"mean_token_accuracy": 0.2244653657078743,
|
|
"num_tokens": 16763370.0,
|
|
"step": 7585
|
|
},
|
|
{
|
|
"entropy": 5.160471868515015,
|
|
"epoch": 7.65153807362582,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00012123154447331064,
|
|
"loss": 4.4318,
|
|
"mean_token_accuracy": 0.2248635396361351,
|
|
"num_tokens": 16772608.0,
|
|
"step": 7590
|
|
},
|
|
{
|
|
"entropy": 5.220185422897339,
|
|
"epoch": 7.656580937972769,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00012094220578747332,
|
|
"loss": 4.4277,
|
|
"mean_token_accuracy": 0.22920091450214386,
|
|
"num_tokens": 16783860.0,
|
|
"step": 7595
|
|
},
|
|
{
|
|
"entropy": 5.185070133209228,
|
|
"epoch": 7.661623802319718,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001206533459168143,
|
|
"loss": 4.5307,
|
|
"mean_token_accuracy": 0.20519898384809493,
|
|
"num_tokens": 16795322.0,
|
|
"step": 7600
|
|
},
|
|
{
|
|
"entropy": 5.167887830734253,
|
|
"epoch": 7.666666666666667,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00012036496575911656,
|
|
"loss": 4.4813,
|
|
"mean_token_accuracy": 0.2190399184823036,
|
|
"num_tokens": 16805873.0,
|
|
"step": 7605
|
|
},
|
|
{
|
|
"entropy": 5.205083608627319,
|
|
"epoch": 7.671709531013616,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00012007706621067232,
|
|
"loss": 4.4644,
|
|
"mean_token_accuracy": 0.22115887999534606,
|
|
"num_tokens": 16816996.0,
|
|
"step": 7610
|
|
},
|
|
{
|
|
"entropy": 5.153089332580566,
|
|
"epoch": 7.676752395360564,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00011978964816628008,
|
|
"loss": 4.4292,
|
|
"mean_token_accuracy": 0.2278457462787628,
|
|
"num_tokens": 16826861.0,
|
|
"step": 7615
|
|
},
|
|
{
|
|
"entropy": 5.2090802669525145,
|
|
"epoch": 7.681795259707513,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00011950271251924178,
|
|
"loss": 4.527,
|
|
"mean_token_accuracy": 0.21956518143415452,
|
|
"num_tokens": 16838533.0,
|
|
"step": 7620
|
|
},
|
|
{
|
|
"entropy": 5.151608943939209,
|
|
"epoch": 7.6868381240544625,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00011921626016135991,
|
|
"loss": 4.4293,
|
|
"mean_token_accuracy": 0.22262226343154906,
|
|
"num_tokens": 16849481.0,
|
|
"step": 7625
|
|
},
|
|
{
|
|
"entropy": 5.058671188354492,
|
|
"epoch": 7.691880988401412,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00011893029198293512,
|
|
"loss": 4.4314,
|
|
"mean_token_accuracy": 0.23021759688854218,
|
|
"num_tokens": 16859388.0,
|
|
"step": 7630
|
|
},
|
|
{
|
|
"entropy": 5.101687240600586,
|
|
"epoch": 7.696923852748361,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00011864480887276301,
|
|
"loss": 4.4509,
|
|
"mean_token_accuracy": 0.22016280591487886,
|
|
"num_tokens": 16870499.0,
|
|
"step": 7635
|
|
},
|
|
{
|
|
"entropy": 5.186880111694336,
|
|
"epoch": 7.70196671709531,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001183598117181317,
|
|
"loss": 4.5383,
|
|
"mean_token_accuracy": 0.21717880219221114,
|
|
"num_tokens": 16880490.0,
|
|
"step": 7640
|
|
},
|
|
{
|
|
"entropy": 5.15349645614624,
|
|
"epoch": 7.707009581442259,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00011807530140481896,
|
|
"loss": 4.5065,
|
|
"mean_token_accuracy": 0.21083239912986756,
|
|
"num_tokens": 16891671.0,
|
|
"step": 7645
|
|
},
|
|
{
|
|
"entropy": 5.074675607681274,
|
|
"epoch": 7.712052445789208,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00011779127881708946,
|
|
"loss": 4.372,
|
|
"mean_token_accuracy": 0.22657344341278077,
|
|
"num_tokens": 16902389.0,
|
|
"step": 7650
|
|
},
|
|
{
|
|
"entropy": 5.101219511032104,
|
|
"epoch": 7.717095310136157,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001175077448376918,
|
|
"loss": 4.4121,
|
|
"mean_token_accuracy": 0.22084709405899047,
|
|
"num_tokens": 16912924.0,
|
|
"step": 7655
|
|
},
|
|
{
|
|
"entropy": 5.166792631149292,
|
|
"epoch": 7.722138174483106,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001172247003478562,
|
|
"loss": 4.5147,
|
|
"mean_token_accuracy": 0.218908853828907,
|
|
"num_tokens": 16923110.0,
|
|
"step": 7660
|
|
},
|
|
{
|
|
"entropy": 5.15488338470459,
|
|
"epoch": 7.7271810388300555,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00011694214622729157,
|
|
"loss": 4.5088,
|
|
"mean_token_accuracy": 0.21795523911714554,
|
|
"num_tokens": 16935034.0,
|
|
"step": 7665
|
|
},
|
|
{
|
|
"entropy": 5.132475900650024,
|
|
"epoch": 7.732223903177005,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00011666008335418241,
|
|
"loss": 4.4712,
|
|
"mean_token_accuracy": 0.21991737931966782,
|
|
"num_tokens": 16945799.0,
|
|
"step": 7670
|
|
},
|
|
{
|
|
"entropy": 5.216502141952515,
|
|
"epoch": 7.737266767523954,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001163785126051868,
|
|
"loss": 4.5073,
|
|
"mean_token_accuracy": 0.2204444468021393,
|
|
"num_tokens": 16957156.0,
|
|
"step": 7675
|
|
},
|
|
{
|
|
"entropy": 5.115852975845337,
|
|
"epoch": 7.742309631870903,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001160974348554331,
|
|
"loss": 4.5,
|
|
"mean_token_accuracy": 0.21736911237239837,
|
|
"num_tokens": 16968439.0,
|
|
"step": 7680
|
|
},
|
|
{
|
|
"entropy": 5.138525104522705,
|
|
"epoch": 7.747352496217852,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00011581685097851747,
|
|
"loss": 4.4958,
|
|
"mean_token_accuracy": 0.21878014653921127,
|
|
"num_tokens": 16979279.0,
|
|
"step": 7685
|
|
},
|
|
{
|
|
"entropy": 5.103876733779908,
|
|
"epoch": 7.752395360564801,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00011553676184650096,
|
|
"loss": 4.4104,
|
|
"mean_token_accuracy": 0.21922295838594436,
|
|
"num_tokens": 16989764.0,
|
|
"step": 7690
|
|
},
|
|
{
|
|
"entropy": 5.168271207809449,
|
|
"epoch": 7.75743822491175,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00011525716832990722,
|
|
"loss": 4.4587,
|
|
"mean_token_accuracy": 0.22370844036340715,
|
|
"num_tokens": 17001355.0,
|
|
"step": 7695
|
|
},
|
|
{
|
|
"entropy": 5.15675835609436,
|
|
"epoch": 7.762481089258699,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00011497807129771916,
|
|
"loss": 4.4932,
|
|
"mean_token_accuracy": 0.21788745671510695,
|
|
"num_tokens": 17012800.0,
|
|
"step": 7700
|
|
},
|
|
{
|
|
"entropy": 5.145964431762695,
|
|
"epoch": 7.767523953605648,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00011469947161737689,
|
|
"loss": 4.5237,
|
|
"mean_token_accuracy": 0.21835473775863648,
|
|
"num_tokens": 17024531.0,
|
|
"step": 7705
|
|
},
|
|
{
|
|
"entropy": 5.188358354568481,
|
|
"epoch": 7.7725668179525975,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00011442137015477464,
|
|
"loss": 4.5073,
|
|
"mean_token_accuracy": 0.2188661739230156,
|
|
"num_tokens": 17035439.0,
|
|
"step": 7710
|
|
},
|
|
{
|
|
"entropy": 5.1519896507263185,
|
|
"epoch": 7.777609682299547,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00011414376777425823,
|
|
"loss": 4.4498,
|
|
"mean_token_accuracy": 0.22127191722393036,
|
|
"num_tokens": 17045831.0,
|
|
"step": 7715
|
|
},
|
|
{
|
|
"entropy": 5.189052963256836,
|
|
"epoch": 7.782652546646495,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00011386666533862214,
|
|
"loss": 4.5731,
|
|
"mean_token_accuracy": 0.21373550295829774,
|
|
"num_tokens": 17057053.0,
|
|
"step": 7720
|
|
},
|
|
{
|
|
"entropy": 5.197714138031006,
|
|
"epoch": 7.787695410993444,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00011359006370910727,
|
|
"loss": 4.5574,
|
|
"mean_token_accuracy": 0.2158846914768219,
|
|
"num_tokens": 17069288.0,
|
|
"step": 7725
|
|
},
|
|
{
|
|
"entropy": 5.203327703475952,
|
|
"epoch": 7.792738275340393,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011331396374539771,
|
|
"loss": 4.5408,
|
|
"mean_token_accuracy": 0.21987006813287735,
|
|
"num_tokens": 17080506.0,
|
|
"step": 7730
|
|
},
|
|
{
|
|
"entropy": 5.1448863506317135,
|
|
"epoch": 7.797781139687342,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00011303836630561861,
|
|
"loss": 4.482,
|
|
"mean_token_accuracy": 0.2191673621535301,
|
|
"num_tokens": 17092483.0,
|
|
"step": 7735
|
|
},
|
|
{
|
|
"entropy": 5.189089393615722,
|
|
"epoch": 7.802824004034291,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00011276327224633311,
|
|
"loss": 4.4797,
|
|
"mean_token_accuracy": 0.21365516781806945,
|
|
"num_tokens": 17103476.0,
|
|
"step": 7740
|
|
},
|
|
{
|
|
"entropy": 5.175717210769653,
|
|
"epoch": 7.80786686838124,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00011248868242254006,
|
|
"loss": 4.4901,
|
|
"mean_token_accuracy": 0.21214143335819244,
|
|
"num_tokens": 17114071.0,
|
|
"step": 7745
|
|
},
|
|
{
|
|
"entropy": 5.1016921043396,
|
|
"epoch": 7.81290973272819,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00011221459768767074,
|
|
"loss": 4.4377,
|
|
"mean_token_accuracy": 0.22491368502378464,
|
|
"num_tokens": 17124989.0,
|
|
"step": 7750
|
|
},
|
|
{
|
|
"entropy": 5.110810375213623,
|
|
"epoch": 7.817952597075139,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00011194101889358702,
|
|
"loss": 4.4014,
|
|
"mean_token_accuracy": 0.22535090446472167,
|
|
"num_tokens": 17135016.0,
|
|
"step": 7755
|
|
},
|
|
{
|
|
"entropy": 5.160736560821533,
|
|
"epoch": 7.822995461422088,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00011166794689057794,
|
|
"loss": 4.4959,
|
|
"mean_token_accuracy": 0.21983225494623185,
|
|
"num_tokens": 17145403.0,
|
|
"step": 7760
|
|
},
|
|
{
|
|
"entropy": 5.1661852359771725,
|
|
"epoch": 7.828038325769037,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00011139538252735765,
|
|
"loss": 4.5167,
|
|
"mean_token_accuracy": 0.21840061843395234,
|
|
"num_tokens": 17156336.0,
|
|
"step": 7765
|
|
},
|
|
{
|
|
"entropy": 5.183116102218628,
|
|
"epoch": 7.833081190115986,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00011112332665106251,
|
|
"loss": 4.5365,
|
|
"mean_token_accuracy": 0.21716483384370805,
|
|
"num_tokens": 17167906.0,
|
|
"step": 7770
|
|
},
|
|
{
|
|
"entropy": 5.094878482818603,
|
|
"epoch": 7.838124054462935,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00011085178010724845,
|
|
"loss": 4.373,
|
|
"mean_token_accuracy": 0.2244934543967247,
|
|
"num_tokens": 17179016.0,
|
|
"step": 7775
|
|
},
|
|
{
|
|
"entropy": 5.199371862411499,
|
|
"epoch": 7.843166918809884,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00011058074373988842,
|
|
"loss": 4.5378,
|
|
"mean_token_accuracy": 0.20928946435451506,
|
|
"num_tokens": 17190656.0,
|
|
"step": 7780
|
|
},
|
|
{
|
|
"entropy": 5.12181282043457,
|
|
"epoch": 7.848209783156833,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001103102183913696,
|
|
"loss": 4.485,
|
|
"mean_token_accuracy": 0.22196391522884368,
|
|
"num_tokens": 17203970.0,
|
|
"step": 7785
|
|
},
|
|
{
|
|
"entropy": 5.167482185363769,
|
|
"epoch": 7.8532526475037825,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00011004020490249116,
|
|
"loss": 4.4865,
|
|
"mean_token_accuracy": 0.22147053033113479,
|
|
"num_tokens": 17215075.0,
|
|
"step": 7790
|
|
},
|
|
{
|
|
"entropy": 5.1579750061035154,
|
|
"epoch": 7.858295511850731,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00010977070411246107,
|
|
"loss": 4.4942,
|
|
"mean_token_accuracy": 0.21468406021595002,
|
|
"num_tokens": 17226065.0,
|
|
"step": 7795
|
|
},
|
|
{
|
|
"entropy": 5.192822122573853,
|
|
"epoch": 7.86333837619768,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00010950171685889413,
|
|
"loss": 4.5342,
|
|
"mean_token_accuracy": 0.22150151878595353,
|
|
"num_tokens": 17238486.0,
|
|
"step": 7800
|
|
},
|
|
{
|
|
"entropy": 5.121309089660644,
|
|
"epoch": 7.868381240544629,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00010923324397780887,
|
|
"loss": 4.5073,
|
|
"mean_token_accuracy": 0.21485970616340638,
|
|
"num_tokens": 17249455.0,
|
|
"step": 7805
|
|
},
|
|
{
|
|
"entropy": 5.124783277511597,
|
|
"epoch": 7.873424104891578,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00010896528630362531,
|
|
"loss": 4.5184,
|
|
"mean_token_accuracy": 0.2175796076655388,
|
|
"num_tokens": 17260311.0,
|
|
"step": 7810
|
|
},
|
|
{
|
|
"entropy": 5.17168927192688,
|
|
"epoch": 7.878466969238527,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010869784466916193,
|
|
"loss": 4.5053,
|
|
"mean_token_accuracy": 0.21560009866952895,
|
|
"num_tokens": 17271551.0,
|
|
"step": 7815
|
|
},
|
|
{
|
|
"entropy": 5.124363994598388,
|
|
"epoch": 7.883509833585476,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00010843091990563365,
|
|
"loss": 4.44,
|
|
"mean_token_accuracy": 0.2221406564116478,
|
|
"num_tokens": 17282401.0,
|
|
"step": 7820
|
|
},
|
|
{
|
|
"entropy": 5.153187799453735,
|
|
"epoch": 7.888552697932425,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00010816451284264864,
|
|
"loss": 4.4315,
|
|
"mean_token_accuracy": 0.22542223930358887,
|
|
"num_tokens": 17293958.0,
|
|
"step": 7825
|
|
},
|
|
{
|
|
"entropy": 5.107201862335205,
|
|
"epoch": 7.8935955622793745,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00010789862430820626,
|
|
"loss": 4.3971,
|
|
"mean_token_accuracy": 0.22622163593769073,
|
|
"num_tokens": 17305342.0,
|
|
"step": 7830
|
|
},
|
|
{
|
|
"entropy": 5.057875537872315,
|
|
"epoch": 7.898638426626324,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00010763325512869422,
|
|
"loss": 4.3588,
|
|
"mean_token_accuracy": 0.2294115349650383,
|
|
"num_tokens": 17315770.0,
|
|
"step": 7835
|
|
},
|
|
{
|
|
"entropy": 5.1447306156158445,
|
|
"epoch": 7.903681290973273,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001073684061288861,
|
|
"loss": 4.5001,
|
|
"mean_token_accuracy": 0.22237968891859056,
|
|
"num_tokens": 17327596.0,
|
|
"step": 7840
|
|
},
|
|
{
|
|
"entropy": 5.087591552734375,
|
|
"epoch": 7.908724155320222,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00010710407813193859,
|
|
"loss": 4.4245,
|
|
"mean_token_accuracy": 0.22255153208971024,
|
|
"num_tokens": 17337951.0,
|
|
"step": 7845
|
|
},
|
|
{
|
|
"entropy": 5.135909223556519,
|
|
"epoch": 7.913767019667171,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00010684027195938936,
|
|
"loss": 4.4928,
|
|
"mean_token_accuracy": 0.2215503066778183,
|
|
"num_tokens": 17348381.0,
|
|
"step": 7850
|
|
},
|
|
{
|
|
"entropy": 5.131261110305786,
|
|
"epoch": 7.91880988401412,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001065769884311539,
|
|
"loss": 4.4334,
|
|
"mean_token_accuracy": 0.21935429871082307,
|
|
"num_tokens": 17357469.0,
|
|
"step": 7855
|
|
},
|
|
{
|
|
"entropy": 5.183663415908813,
|
|
"epoch": 7.923852748361069,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00010631422836552354,
|
|
"loss": 4.4668,
|
|
"mean_token_accuracy": 0.2217352107167244,
|
|
"num_tokens": 17368469.0,
|
|
"step": 7860
|
|
},
|
|
{
|
|
"entropy": 5.177272796630859,
|
|
"epoch": 7.928895612708018,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00010605199257916269,
|
|
"loss": 4.4678,
|
|
"mean_token_accuracy": 0.21930347383022308,
|
|
"num_tokens": 17379378.0,
|
|
"step": 7865
|
|
},
|
|
{
|
|
"entropy": 5.1260254859924315,
|
|
"epoch": 7.9339384770549675,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00010579028188710619,
|
|
"loss": 4.4748,
|
|
"mean_token_accuracy": 0.2189782127737999,
|
|
"num_tokens": 17391449.0,
|
|
"step": 7870
|
|
},
|
|
{
|
|
"entropy": 5.141279506683349,
|
|
"epoch": 7.938981341401917,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00010552909710275693,
|
|
"loss": 4.5141,
|
|
"mean_token_accuracy": 0.21308174282312392,
|
|
"num_tokens": 17401360.0,
|
|
"step": 7875
|
|
},
|
|
{
|
|
"entropy": 5.124091100692749,
|
|
"epoch": 7.944024205748866,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00010526843903788319,
|
|
"loss": 4.443,
|
|
"mean_token_accuracy": 0.22832788228988649,
|
|
"num_tokens": 17411367.0,
|
|
"step": 7880
|
|
},
|
|
{
|
|
"entropy": 5.217112493515015,
|
|
"epoch": 7.949067070095815,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00010500830850261634,
|
|
"loss": 4.5531,
|
|
"mean_token_accuracy": 0.2208593085408211,
|
|
"num_tokens": 17421920.0,
|
|
"step": 7885
|
|
},
|
|
{
|
|
"entropy": 5.173935508728027,
|
|
"epoch": 7.954109934442764,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00010474870630544797,
|
|
"loss": 4.4831,
|
|
"mean_token_accuracy": 0.22745779156684875,
|
|
"num_tokens": 17432843.0,
|
|
"step": 7890
|
|
},
|
|
{
|
|
"entropy": 5.137723970413208,
|
|
"epoch": 7.959152798789713,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00010448963325322772,
|
|
"loss": 4.436,
|
|
"mean_token_accuracy": 0.22295926809310912,
|
|
"num_tokens": 17443445.0,
|
|
"step": 7895
|
|
},
|
|
{
|
|
"entropy": 5.201297235488892,
|
|
"epoch": 7.964195663136661,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00010423109015116065,
|
|
"loss": 4.5423,
|
|
"mean_token_accuracy": 0.21130658388137818,
|
|
"num_tokens": 17454956.0,
|
|
"step": 7900
|
|
},
|
|
{
|
|
"entropy": 5.204122686386109,
|
|
"epoch": 7.96923852748361,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00010397307780280477,
|
|
"loss": 4.5212,
|
|
"mean_token_accuracy": 0.21425657719373703,
|
|
"num_tokens": 17466874.0,
|
|
"step": 7905
|
|
},
|
|
{
|
|
"entropy": 5.1651064395904545,
|
|
"epoch": 7.9742813918305595,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00010371559701006825,
|
|
"loss": 4.5022,
|
|
"mean_token_accuracy": 0.2220512554049492,
|
|
"num_tokens": 17477194.0,
|
|
"step": 7910
|
|
},
|
|
{
|
|
"entropy": 5.12298994064331,
|
|
"epoch": 7.979324256177509,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00010345864857320747,
|
|
"loss": 4.4284,
|
|
"mean_token_accuracy": 0.22302357703447342,
|
|
"num_tokens": 17488115.0,
|
|
"step": 7915
|
|
},
|
|
{
|
|
"entropy": 5.139308214187622,
|
|
"epoch": 7.984367120524458,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00010320223329082396,
|
|
"loss": 4.4225,
|
|
"mean_token_accuracy": 0.22407811135053635,
|
|
"num_tokens": 17498354.0,
|
|
"step": 7920
|
|
},
|
|
{
|
|
"entropy": 5.068280553817749,
|
|
"epoch": 7.989409984871407,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001029463519598624,
|
|
"loss": 4.4129,
|
|
"mean_token_accuracy": 0.22646554112434386,
|
|
"num_tokens": 17509953.0,
|
|
"step": 7925
|
|
},
|
|
{
|
|
"entropy": 5.07424144744873,
|
|
"epoch": 7.994452849218356,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00010269100537560789,
|
|
"loss": 4.4275,
|
|
"mean_token_accuracy": 0.22144773304462434,
|
|
"num_tokens": 17522229.0,
|
|
"step": 7930
|
|
},
|
|
{
|
|
"entropy": 5.138594675064087,
|
|
"epoch": 7.999495713565305,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010243619433168352,
|
|
"loss": 4.4737,
|
|
"mean_token_accuracy": 0.2225889101624489,
|
|
"num_tokens": 17532528.0,
|
|
"step": 7935
|
|
},
|
|
{
|
|
"entropy": 5.187544504801433,
|
|
"epoch": 8.00403429147756,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010218191962004781,
|
|
"loss": 4.4272,
|
|
"mean_token_accuracy": 0.22841538820001814,
|
|
"num_tokens": 17540957.0,
|
|
"step": 7940
|
|
},
|
|
{
|
|
"entropy": 5.182134103775025,
|
|
"epoch": 8.009077155824508,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00010192818203099258,
|
|
"loss": 4.3852,
|
|
"mean_token_accuracy": 0.22523926347494125,
|
|
"num_tokens": 17551542.0,
|
|
"step": 7945
|
|
},
|
|
{
|
|
"entropy": 5.106996202468872,
|
|
"epoch": 8.014120020171458,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001016749823531399,
|
|
"loss": 4.352,
|
|
"mean_token_accuracy": 0.23374103903770446,
|
|
"num_tokens": 17562709.0,
|
|
"step": 7950
|
|
},
|
|
{
|
|
"entropy": 5.122921514511108,
|
|
"epoch": 8.019162884518407,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00010142232137344036,
|
|
"loss": 4.3287,
|
|
"mean_token_accuracy": 0.23751920014619826,
|
|
"num_tokens": 17573939.0,
|
|
"step": 7955
|
|
},
|
|
{
|
|
"entropy": 5.150394773483276,
|
|
"epoch": 8.024205748865356,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00010117019987717005,
|
|
"loss": 4.3683,
|
|
"mean_token_accuracy": 0.22924542427062988,
|
|
"num_tokens": 17584403.0,
|
|
"step": 7960
|
|
},
|
|
{
|
|
"entropy": 5.133014869689942,
|
|
"epoch": 8.029248613212305,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00010091861864792848,
|
|
"loss": 4.3799,
|
|
"mean_token_accuracy": 0.2309165731072426,
|
|
"num_tokens": 17594191.0,
|
|
"step": 7965
|
|
},
|
|
{
|
|
"entropy": 5.168467903137207,
|
|
"epoch": 8.034291477559254,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001006675784676358,
|
|
"loss": 4.5331,
|
|
"mean_token_accuracy": 0.2160644456744194,
|
|
"num_tokens": 17605525.0,
|
|
"step": 7970
|
|
},
|
|
{
|
|
"entropy": 5.093361473083496,
|
|
"epoch": 8.039334341906203,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00010041708011653081,
|
|
"loss": 4.3179,
|
|
"mean_token_accuracy": 0.23210108876228333,
|
|
"num_tokens": 17616683.0,
|
|
"step": 7975
|
|
},
|
|
{
|
|
"entropy": 5.088868474960327,
|
|
"epoch": 8.044377206253152,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00010016712437316799,
|
|
"loss": 4.3216,
|
|
"mean_token_accuracy": 0.23262573927640914,
|
|
"num_tokens": 17627444.0,
|
|
"step": 7980
|
|
},
|
|
{
|
|
"entropy": 5.141922235488892,
|
|
"epoch": 8.049420070600101,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 9.991771201441569e-05,
|
|
"loss": 4.4231,
|
|
"mean_token_accuracy": 0.22620819211006166,
|
|
"num_tokens": 17637756.0,
|
|
"step": 7985
|
|
},
|
|
{
|
|
"entropy": 5.120128154754639,
|
|
"epoch": 8.05446293494705,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 9.966884381545322e-05,
|
|
"loss": 4.3389,
|
|
"mean_token_accuracy": 0.23265717625617982,
|
|
"num_tokens": 17649423.0,
|
|
"step": 7990
|
|
},
|
|
{
|
|
"entropy": 5.1354917049407955,
|
|
"epoch": 8.059505799294,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 9.942052054976872e-05,
|
|
"loss": 4.3642,
|
|
"mean_token_accuracy": 0.23446649461984634,
|
|
"num_tokens": 17661295.0,
|
|
"step": 7995
|
|
},
|
|
{
|
|
"entropy": 5.211695384979248,
|
|
"epoch": 8.064548663640949,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.917274298915665e-05,
|
|
"loss": 4.3971,
|
|
"mean_token_accuracy": 0.2287270039319992,
|
|
"num_tokens": 17672189.0,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"epoch": 8.064548663640949,
|
|
"eval_entropy": 5.114620173553777,
|
|
"eval_loss": 5.698058605194092,
|
|
"eval_mean_token_accuracy": 0.15586839493070243,
|
|
"eval_num_tokens": 17672189.0,
|
|
"eval_runtime": 2.6435,
|
|
"eval_samples_per_second": 1187.044,
|
|
"eval_steps_per_second": 148.664,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"entropy": 5.158815574645996,
|
|
"epoch": 8.069591527987898,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 9.892551190371529e-05,
|
|
"loss": 4.3715,
|
|
"mean_token_accuracy": 0.2250150129199028,
|
|
"num_tokens": 17683919.0,
|
|
"step": 8005
|
|
},
|
|
{
|
|
"entropy": 5.111160516738892,
|
|
"epoch": 8.074634392334847,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 9.867882806184469e-05,
|
|
"loss": 4.372,
|
|
"mean_token_accuracy": 0.23340094238519668,
|
|
"num_tokens": 17694974.0,
|
|
"step": 8010
|
|
},
|
|
{
|
|
"entropy": 5.116820001602173,
|
|
"epoch": 8.079677256681796,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 9.843269223024376e-05,
|
|
"loss": 4.3948,
|
|
"mean_token_accuracy": 0.22381257116794587,
|
|
"num_tokens": 17705917.0,
|
|
"step": 8015
|
|
},
|
|
{
|
|
"entropy": 5.128374814987183,
|
|
"epoch": 8.084720121028745,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 9.818710517390841e-05,
|
|
"loss": 4.4375,
|
|
"mean_token_accuracy": 0.22786244451999665,
|
|
"num_tokens": 17717842.0,
|
|
"step": 8020
|
|
},
|
|
{
|
|
"entropy": 5.157124853134155,
|
|
"epoch": 8.089762985375694,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 9.794206765612893e-05,
|
|
"loss": 4.4368,
|
|
"mean_token_accuracy": 0.23023964911699296,
|
|
"num_tokens": 17730497.0,
|
|
"step": 8025
|
|
},
|
|
{
|
|
"entropy": 5.195124626159668,
|
|
"epoch": 8.094805849722642,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 9.76975804384876e-05,
|
|
"loss": 4.449,
|
|
"mean_token_accuracy": 0.2241278663277626,
|
|
"num_tokens": 17742963.0,
|
|
"step": 8030
|
|
},
|
|
{
|
|
"entropy": 5.170260906219482,
|
|
"epoch": 8.09984871406959,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 9.745364428085618e-05,
|
|
"loss": 4.3811,
|
|
"mean_token_accuracy": 0.22226705253124238,
|
|
"num_tokens": 17754648.0,
|
|
"step": 8035
|
|
},
|
|
{
|
|
"entropy": 5.157980251312256,
|
|
"epoch": 8.10489157841654,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.721025994139404e-05,
|
|
"loss": 4.3855,
|
|
"mean_token_accuracy": 0.22963685393333436,
|
|
"num_tokens": 17764976.0,
|
|
"step": 8040
|
|
},
|
|
{
|
|
"entropy": 5.168565320968628,
|
|
"epoch": 8.109934442763489,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 9.696742817654519e-05,
|
|
"loss": 4.3643,
|
|
"mean_token_accuracy": 0.23790484219789504,
|
|
"num_tokens": 17776787.0,
|
|
"step": 8045
|
|
},
|
|
{
|
|
"entropy": 5.045006608963012,
|
|
"epoch": 8.114977307110438,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 9.672514974103643e-05,
|
|
"loss": 4.2694,
|
|
"mean_token_accuracy": 0.24045135080814362,
|
|
"num_tokens": 17787639.0,
|
|
"step": 8050
|
|
},
|
|
{
|
|
"entropy": 5.121422672271729,
|
|
"epoch": 8.120020171457387,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 9.648342538787467e-05,
|
|
"loss": 4.342,
|
|
"mean_token_accuracy": 0.22763330936431886,
|
|
"num_tokens": 17797246.0,
|
|
"step": 8055
|
|
},
|
|
{
|
|
"entropy": 5.127322149276734,
|
|
"epoch": 8.125063035804336,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 9.62422558683449e-05,
|
|
"loss": 4.4168,
|
|
"mean_token_accuracy": 0.21887322515249252,
|
|
"num_tokens": 17809096.0,
|
|
"step": 8060
|
|
},
|
|
{
|
|
"entropy": 5.1505333423614506,
|
|
"epoch": 8.130105900151285,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 9.600164193200742e-05,
|
|
"loss": 4.4312,
|
|
"mean_token_accuracy": 0.2234543025493622,
|
|
"num_tokens": 17820740.0,
|
|
"step": 8065
|
|
},
|
|
{
|
|
"entropy": 5.141790771484375,
|
|
"epoch": 8.135148764498235,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 9.576158432669599e-05,
|
|
"loss": 4.3518,
|
|
"mean_token_accuracy": 0.22968435734510423,
|
|
"num_tokens": 17832220.0,
|
|
"step": 8070
|
|
},
|
|
{
|
|
"entropy": 5.181501007080078,
|
|
"epoch": 8.140191628845184,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 9.552208379851507e-05,
|
|
"loss": 4.3758,
|
|
"mean_token_accuracy": 0.22861322164535522,
|
|
"num_tokens": 17843597.0,
|
|
"step": 8075
|
|
},
|
|
{
|
|
"entropy": 5.132485055923462,
|
|
"epoch": 8.145234493192133,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 9.528314109183788e-05,
|
|
"loss": 4.3642,
|
|
"mean_token_accuracy": 0.22668164372444152,
|
|
"num_tokens": 17855604.0,
|
|
"step": 8080
|
|
},
|
|
{
|
|
"entropy": 5.131725835800171,
|
|
"epoch": 8.150277357539082,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 9.504475694930392e-05,
|
|
"loss": 4.3739,
|
|
"mean_token_accuracy": 0.22892633378505706,
|
|
"num_tokens": 17866956.0,
|
|
"step": 8085
|
|
},
|
|
{
|
|
"entropy": 5.101851272583008,
|
|
"epoch": 8.155320221886031,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 9.480693211181654e-05,
|
|
"loss": 4.3483,
|
|
"mean_token_accuracy": 0.23558836281299592,
|
|
"num_tokens": 17878358.0,
|
|
"step": 8090
|
|
},
|
|
{
|
|
"entropy": 5.164977407455444,
|
|
"epoch": 8.16036308623298,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.456966731854089e-05,
|
|
"loss": 4.4114,
|
|
"mean_token_accuracy": 0.2252320408821106,
|
|
"num_tokens": 17888229.0,
|
|
"step": 8095
|
|
},
|
|
{
|
|
"entropy": 5.141616249084473,
|
|
"epoch": 8.16540595057993,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.43329633069013e-05,
|
|
"loss": 4.3525,
|
|
"mean_token_accuracy": 0.23040709644556046,
|
|
"num_tokens": 17898051.0,
|
|
"step": 8100
|
|
},
|
|
{
|
|
"entropy": 5.13908748626709,
|
|
"epoch": 8.170448814926878,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 9.409682081257947e-05,
|
|
"loss": 4.3491,
|
|
"mean_token_accuracy": 0.2316196858882904,
|
|
"num_tokens": 17909532.0,
|
|
"step": 8105
|
|
},
|
|
{
|
|
"entropy": 5.199942874908447,
|
|
"epoch": 8.175491679273827,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 9.386124056951154e-05,
|
|
"loss": 4.5153,
|
|
"mean_token_accuracy": 0.21589007526636123,
|
|
"num_tokens": 17921509.0,
|
|
"step": 8110
|
|
},
|
|
{
|
|
"entropy": 5.133984565734863,
|
|
"epoch": 8.180534543620777,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.362622330988641e-05,
|
|
"loss": 4.3813,
|
|
"mean_token_accuracy": 0.2245675340294838,
|
|
"num_tokens": 17931196.0,
|
|
"step": 8115
|
|
},
|
|
{
|
|
"entropy": 5.172988176345825,
|
|
"epoch": 8.185577407967726,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 9.339176976414317e-05,
|
|
"loss": 4.4439,
|
|
"mean_token_accuracy": 0.22419338226318358,
|
|
"num_tokens": 17942327.0,
|
|
"step": 8120
|
|
},
|
|
{
|
|
"entropy": 5.133330917358398,
|
|
"epoch": 8.190620272314675,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.315788066096892e-05,
|
|
"loss": 4.3283,
|
|
"mean_token_accuracy": 0.2402695581316948,
|
|
"num_tokens": 17952361.0,
|
|
"step": 8125
|
|
},
|
|
{
|
|
"entropy": 5.087899827957154,
|
|
"epoch": 8.195663136661624,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 9.292455672729619e-05,
|
|
"loss": 4.3064,
|
|
"mean_token_accuracy": 0.22956128120422364,
|
|
"num_tokens": 17962732.0,
|
|
"step": 8130
|
|
},
|
|
{
|
|
"entropy": 5.135718870162964,
|
|
"epoch": 8.200706001008573,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.269179868830135e-05,
|
|
"loss": 4.4305,
|
|
"mean_token_accuracy": 0.21926121264696122,
|
|
"num_tokens": 17972036.0,
|
|
"step": 8135
|
|
},
|
|
{
|
|
"entropy": 5.107414245605469,
|
|
"epoch": 8.205748865355522,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 9.24596072674016e-05,
|
|
"loss": 4.4421,
|
|
"mean_token_accuracy": 0.225291445851326,
|
|
"num_tokens": 17984349.0,
|
|
"step": 8140
|
|
},
|
|
{
|
|
"entropy": 5.096823263168335,
|
|
"epoch": 8.210791729702471,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 9.222798318625327e-05,
|
|
"loss": 4.306,
|
|
"mean_token_accuracy": 0.23327143043279647,
|
|
"num_tokens": 17995187.0,
|
|
"step": 8145
|
|
},
|
|
{
|
|
"entropy": 5.131308746337891,
|
|
"epoch": 8.21583459404942,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 9.19969271647494e-05,
|
|
"loss": 4.3132,
|
|
"mean_token_accuracy": 0.23628715872764589,
|
|
"num_tokens": 18005444.0,
|
|
"step": 8150
|
|
},
|
|
{
|
|
"entropy": 5.140311813354492,
|
|
"epoch": 8.22087745839637,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 9.176643992101749e-05,
|
|
"loss": 4.3744,
|
|
"mean_token_accuracy": 0.22880188524723052,
|
|
"num_tokens": 18016966.0,
|
|
"step": 8155
|
|
},
|
|
{
|
|
"entropy": 5.119467878341675,
|
|
"epoch": 8.225920322743319,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 9.153652217141703e-05,
|
|
"loss": 4.4412,
|
|
"mean_token_accuracy": 0.22089037895202637,
|
|
"num_tokens": 18028917.0,
|
|
"step": 8160
|
|
},
|
|
{
|
|
"entropy": 5.148067045211792,
|
|
"epoch": 8.230963187090268,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 9.130717463053787e-05,
|
|
"loss": 4.362,
|
|
"mean_token_accuracy": 0.23194685727357864,
|
|
"num_tokens": 18039500.0,
|
|
"step": 8165
|
|
},
|
|
{
|
|
"entropy": 5.133250284194946,
|
|
"epoch": 8.236006051437217,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 9.10783980111973e-05,
|
|
"loss": 4.3481,
|
|
"mean_token_accuracy": 0.2277764305472374,
|
|
"num_tokens": 18048793.0,
|
|
"step": 8170
|
|
},
|
|
{
|
|
"entropy": 5.192077016830444,
|
|
"epoch": 8.241048915784166,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.085019302443842e-05,
|
|
"loss": 4.4416,
|
|
"mean_token_accuracy": 0.22600506693124772,
|
|
"num_tokens": 18060269.0,
|
|
"step": 8175
|
|
},
|
|
{
|
|
"entropy": 5.189511728286743,
|
|
"epoch": 8.246091780131115,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 9.062256037952757e-05,
|
|
"loss": 4.4212,
|
|
"mean_token_accuracy": 0.22864135056734086,
|
|
"num_tokens": 18071488.0,
|
|
"step": 8180
|
|
},
|
|
{
|
|
"entropy": 5.133086204528809,
|
|
"epoch": 8.251134644478064,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 9.039550078395234e-05,
|
|
"loss": 4.4018,
|
|
"mean_token_accuracy": 0.2289685070514679,
|
|
"num_tokens": 18083016.0,
|
|
"step": 8185
|
|
},
|
|
{
|
|
"entropy": 5.241964197158813,
|
|
"epoch": 8.256177508825013,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 9.016901494341906e-05,
|
|
"loss": 4.4908,
|
|
"mean_token_accuracy": 0.21302641928195953,
|
|
"num_tokens": 18095744.0,
|
|
"step": 8190
|
|
},
|
|
{
|
|
"entropy": 5.12712869644165,
|
|
"epoch": 8.261220373171962,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 8.994310356185098e-05,
|
|
"loss": 4.4352,
|
|
"mean_token_accuracy": 0.22613762468099594,
|
|
"num_tokens": 18108409.0,
|
|
"step": 8195
|
|
},
|
|
{
|
|
"entropy": 5.138139486312866,
|
|
"epoch": 8.266263237518912,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 8.971776734138601e-05,
|
|
"loss": 4.3923,
|
|
"mean_token_accuracy": 0.22301608622074126,
|
|
"num_tokens": 18118441.0,
|
|
"step": 8200
|
|
},
|
|
{
|
|
"entropy": 5.19263277053833,
|
|
"epoch": 8.27130610186586,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.949300698237414e-05,
|
|
"loss": 4.408,
|
|
"mean_token_accuracy": 0.2252144619822502,
|
|
"num_tokens": 18130686.0,
|
|
"step": 8205
|
|
},
|
|
{
|
|
"entropy": 5.105653858184814,
|
|
"epoch": 8.276348966212808,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.926882318337586e-05,
|
|
"loss": 4.347,
|
|
"mean_token_accuracy": 0.2319524481892586,
|
|
"num_tokens": 18141361.0,
|
|
"step": 8210
|
|
},
|
|
{
|
|
"entropy": 5.123333740234375,
|
|
"epoch": 8.281391830559757,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 8.904521664115956e-05,
|
|
"loss": 4.4098,
|
|
"mean_token_accuracy": 0.2209177315235138,
|
|
"num_tokens": 18154000.0,
|
|
"step": 8215
|
|
},
|
|
{
|
|
"entropy": 5.103848314285278,
|
|
"epoch": 8.286434694906706,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 8.882218805069958e-05,
|
|
"loss": 4.4223,
|
|
"mean_token_accuracy": 0.22132879197597505,
|
|
"num_tokens": 18164516.0,
|
|
"step": 8220
|
|
},
|
|
{
|
|
"entropy": 5.135469532012939,
|
|
"epoch": 8.291477559253655,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 8.859973810517383e-05,
|
|
"loss": 4.3744,
|
|
"mean_token_accuracy": 0.22969135642051697,
|
|
"num_tokens": 18175756.0,
|
|
"step": 8225
|
|
},
|
|
{
|
|
"entropy": 5.065666961669922,
|
|
"epoch": 8.296520423600604,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 8.8377867495962e-05,
|
|
"loss": 4.392,
|
|
"mean_token_accuracy": 0.22354791164398194,
|
|
"num_tokens": 18187157.0,
|
|
"step": 8230
|
|
},
|
|
{
|
|
"entropy": 5.175729608535766,
|
|
"epoch": 8.301563287947554,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 8.815657691264288e-05,
|
|
"loss": 4.451,
|
|
"mean_token_accuracy": 0.22259393334388733,
|
|
"num_tokens": 18197186.0,
|
|
"step": 8235
|
|
},
|
|
{
|
|
"entropy": 5.088168096542359,
|
|
"epoch": 8.306606152294503,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 8.793586704299284e-05,
|
|
"loss": 4.3757,
|
|
"mean_token_accuracy": 0.22903859466314316,
|
|
"num_tokens": 18208932.0,
|
|
"step": 8240
|
|
},
|
|
{
|
|
"entropy": 5.131833410263061,
|
|
"epoch": 8.311649016641452,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.77157385729832e-05,
|
|
"loss": 4.4477,
|
|
"mean_token_accuracy": 0.22275378108024596,
|
|
"num_tokens": 18220208.0,
|
|
"step": 8245
|
|
},
|
|
{
|
|
"entropy": 5.100909757614136,
|
|
"epoch": 8.316691880988401,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 8.749619218677842e-05,
|
|
"loss": 4.4112,
|
|
"mean_token_accuracy": 0.227280493080616,
|
|
"num_tokens": 18231681.0,
|
|
"step": 8250
|
|
},
|
|
{
|
|
"entropy": 5.175965976715088,
|
|
"epoch": 8.32173474533535,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.727722856673357e-05,
|
|
"loss": 4.4313,
|
|
"mean_token_accuracy": 0.22667446732521057,
|
|
"num_tokens": 18242466.0,
|
|
"step": 8255
|
|
},
|
|
{
|
|
"entropy": 5.21811056137085,
|
|
"epoch": 8.3267776096823,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 8.705884839339279e-05,
|
|
"loss": 4.4556,
|
|
"mean_token_accuracy": 0.22411052733659745,
|
|
"num_tokens": 18254451.0,
|
|
"step": 8260
|
|
},
|
|
{
|
|
"entropy": 5.190957689285279,
|
|
"epoch": 8.331820474029248,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 8.684105234548658e-05,
|
|
"loss": 4.4564,
|
|
"mean_token_accuracy": 0.22370578199625016,
|
|
"num_tokens": 18265285.0,
|
|
"step": 8265
|
|
},
|
|
{
|
|
"entropy": 5.235081434249878,
|
|
"epoch": 8.336863338376197,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.66238410999301e-05,
|
|
"loss": 4.5224,
|
|
"mean_token_accuracy": 0.21813571006059645,
|
|
"num_tokens": 18277188.0,
|
|
"step": 8270
|
|
},
|
|
{
|
|
"entropy": 5.192430686950684,
|
|
"epoch": 8.341906202723147,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 8.640721533182092e-05,
|
|
"loss": 4.455,
|
|
"mean_token_accuracy": 0.21539300680160522,
|
|
"num_tokens": 18288377.0,
|
|
"step": 8275
|
|
},
|
|
{
|
|
"entropy": 5.1049939632415775,
|
|
"epoch": 8.346949067070096,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.619117571443702e-05,
|
|
"loss": 4.352,
|
|
"mean_token_accuracy": 0.24022766202688217,
|
|
"num_tokens": 18299798.0,
|
|
"step": 8280
|
|
},
|
|
{
|
|
"entropy": 5.152305459976196,
|
|
"epoch": 8.351991931417045,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 8.597572291923438e-05,
|
|
"loss": 4.4005,
|
|
"mean_token_accuracy": 0.22155966758728027,
|
|
"num_tokens": 18310382.0,
|
|
"step": 8285
|
|
},
|
|
{
|
|
"entropy": 5.110679197311401,
|
|
"epoch": 8.357034795763994,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 8.576085761584534e-05,
|
|
"loss": 4.3519,
|
|
"mean_token_accuracy": 0.2328895092010498,
|
|
"num_tokens": 18321200.0,
|
|
"step": 8290
|
|
},
|
|
{
|
|
"entropy": 5.143840885162353,
|
|
"epoch": 8.362077660110943,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.554658047207616e-05,
|
|
"loss": 4.4814,
|
|
"mean_token_accuracy": 0.2250434100627899,
|
|
"num_tokens": 18331885.0,
|
|
"step": 8295
|
|
},
|
|
{
|
|
"entropy": 5.125862550735474,
|
|
"epoch": 8.367120524457892,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 8.533289215390516e-05,
|
|
"loss": 4.3953,
|
|
"mean_token_accuracy": 0.23426084518432616,
|
|
"num_tokens": 18341448.0,
|
|
"step": 8300
|
|
},
|
|
{
|
|
"entropy": 5.145696401596069,
|
|
"epoch": 8.372163388804841,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 8.51197933254806e-05,
|
|
"loss": 4.4111,
|
|
"mean_token_accuracy": 0.22590701431035995,
|
|
"num_tokens": 18352737.0,
|
|
"step": 8305
|
|
},
|
|
{
|
|
"entropy": 5.146053981781006,
|
|
"epoch": 8.37720625315179,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 8.490728464911849e-05,
|
|
"loss": 4.413,
|
|
"mean_token_accuracy": 0.22714868187904358,
|
|
"num_tokens": 18362354.0,
|
|
"step": 8310
|
|
},
|
|
{
|
|
"entropy": 5.1328212261199955,
|
|
"epoch": 8.38224911749874,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.469536678530085e-05,
|
|
"loss": 4.4155,
|
|
"mean_token_accuracy": 0.22540309578180312,
|
|
"num_tokens": 18373436.0,
|
|
"step": 8315
|
|
},
|
|
{
|
|
"entropy": 5.132914733886719,
|
|
"epoch": 8.387291981845689,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 8.448404039267308e-05,
|
|
"loss": 4.362,
|
|
"mean_token_accuracy": 0.2297257050871849,
|
|
"num_tokens": 18383808.0,
|
|
"step": 8320
|
|
},
|
|
{
|
|
"entropy": 5.144194507598877,
|
|
"epoch": 8.392334846192638,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 8.427330612804266e-05,
|
|
"loss": 4.451,
|
|
"mean_token_accuracy": 0.22346042096614838,
|
|
"num_tokens": 18395214.0,
|
|
"step": 8325
|
|
},
|
|
{
|
|
"entropy": 5.197042560577392,
|
|
"epoch": 8.397377710539587,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.406316464637639e-05,
|
|
"loss": 4.3608,
|
|
"mean_token_accuracy": 0.22875447571277618,
|
|
"num_tokens": 18405477.0,
|
|
"step": 8330
|
|
},
|
|
{
|
|
"entropy": 5.12706446647644,
|
|
"epoch": 8.402420574886536,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.385361660079896e-05,
|
|
"loss": 4.3519,
|
|
"mean_token_accuracy": 0.23964355289936065,
|
|
"num_tokens": 18415195.0,
|
|
"step": 8335
|
|
},
|
|
{
|
|
"entropy": 5.14655499458313,
|
|
"epoch": 8.407463439233485,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 8.364466264259046e-05,
|
|
"loss": 4.4385,
|
|
"mean_token_accuracy": 0.2261045604944229,
|
|
"num_tokens": 18426043.0,
|
|
"step": 8340
|
|
},
|
|
{
|
|
"entropy": 5.173324060440064,
|
|
"epoch": 8.412506303580434,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 8.343630342118471e-05,
|
|
"loss": 4.4615,
|
|
"mean_token_accuracy": 0.2198323428630829,
|
|
"num_tokens": 18436659.0,
|
|
"step": 8345
|
|
},
|
|
{
|
|
"entropy": 5.13218502998352,
|
|
"epoch": 8.417549167927383,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 8.322853958416685e-05,
|
|
"loss": 4.3821,
|
|
"mean_token_accuracy": 0.2277967318892479,
|
|
"num_tokens": 18446294.0,
|
|
"step": 8350
|
|
},
|
|
{
|
|
"entropy": 5.120608615875244,
|
|
"epoch": 8.422592032274332,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 8.30213717772718e-05,
|
|
"loss": 4.3712,
|
|
"mean_token_accuracy": 0.23082223683595657,
|
|
"num_tokens": 18457565.0,
|
|
"step": 8355
|
|
},
|
|
{
|
|
"entropy": 5.217173147201538,
|
|
"epoch": 8.427634896621282,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 8.281480064438176e-05,
|
|
"loss": 4.4464,
|
|
"mean_token_accuracy": 0.2243775650858879,
|
|
"num_tokens": 18469267.0,
|
|
"step": 8360
|
|
},
|
|
{
|
|
"entropy": 5.1128252983093265,
|
|
"epoch": 8.43267776096823,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 8.260882682752464e-05,
|
|
"loss": 4.4054,
|
|
"mean_token_accuracy": 0.23256094604730607,
|
|
"num_tokens": 18481937.0,
|
|
"step": 8365
|
|
},
|
|
{
|
|
"entropy": 5.160320806503296,
|
|
"epoch": 8.43772062531518,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 8.24034509668718e-05,
|
|
"loss": 4.4925,
|
|
"mean_token_accuracy": 0.21604703217744828,
|
|
"num_tokens": 18494431.0,
|
|
"step": 8370
|
|
},
|
|
{
|
|
"entropy": 5.116606044769287,
|
|
"epoch": 8.442763489662129,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 8.219867370073623e-05,
|
|
"loss": 4.3911,
|
|
"mean_token_accuracy": 0.22473449558019637,
|
|
"num_tokens": 18504692.0,
|
|
"step": 8375
|
|
},
|
|
{
|
|
"entropy": 5.189106178283692,
|
|
"epoch": 8.447806354009078,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 8.199449566557029e-05,
|
|
"loss": 4.4193,
|
|
"mean_token_accuracy": 0.22790949195623397,
|
|
"num_tokens": 18515774.0,
|
|
"step": 8380
|
|
},
|
|
{
|
|
"entropy": 5.151889371871948,
|
|
"epoch": 8.452849218356025,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 8.179091749596409e-05,
|
|
"loss": 4.4231,
|
|
"mean_token_accuracy": 0.2202611342072487,
|
|
"num_tokens": 18526603.0,
|
|
"step": 8385
|
|
},
|
|
{
|
|
"entropy": 5.151511526107788,
|
|
"epoch": 8.457892082702974,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 8.158793982464322e-05,
|
|
"loss": 4.4374,
|
|
"mean_token_accuracy": 0.22261128425598145,
|
|
"num_tokens": 18538021.0,
|
|
"step": 8390
|
|
},
|
|
{
|
|
"entropy": 5.165116310119629,
|
|
"epoch": 8.462934947049924,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 8.138556328246698e-05,
|
|
"loss": 4.4364,
|
|
"mean_token_accuracy": 0.21975694745779037,
|
|
"num_tokens": 18548277.0,
|
|
"step": 8395
|
|
},
|
|
{
|
|
"entropy": 5.1384986400604244,
|
|
"epoch": 8.467977811396873,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 8.118378849842634e-05,
|
|
"loss": 4.4137,
|
|
"mean_token_accuracy": 0.22231665700674058,
|
|
"num_tokens": 18559334.0,
|
|
"step": 8400
|
|
},
|
|
{
|
|
"entropy": 5.172024440765381,
|
|
"epoch": 8.473020675743822,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 8.098261609964204e-05,
|
|
"loss": 4.3911,
|
|
"mean_token_accuracy": 0.22618340700864792,
|
|
"num_tokens": 18570784.0,
|
|
"step": 8405
|
|
},
|
|
{
|
|
"entropy": 5.1844954013824465,
|
|
"epoch": 8.478063540090771,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 8.078204671136245e-05,
|
|
"loss": 4.4035,
|
|
"mean_token_accuracy": 0.22604216635227203,
|
|
"num_tokens": 18582147.0,
|
|
"step": 8410
|
|
},
|
|
{
|
|
"entropy": 5.220875310897827,
|
|
"epoch": 8.48310640443772,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 8.05820809569619e-05,
|
|
"loss": 4.5286,
|
|
"mean_token_accuracy": 0.219061741232872,
|
|
"num_tokens": 18594753.0,
|
|
"step": 8415
|
|
},
|
|
{
|
|
"entropy": 5.180904293060303,
|
|
"epoch": 8.48814926878467,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.03827194579386e-05,
|
|
"loss": 4.4205,
|
|
"mean_token_accuracy": 0.22993909567594528,
|
|
"num_tokens": 18605631.0,
|
|
"step": 8420
|
|
},
|
|
{
|
|
"entropy": 5.1352697849273685,
|
|
"epoch": 8.493192133131618,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 8.01839628339126e-05,
|
|
"loss": 4.3984,
|
|
"mean_token_accuracy": 0.22344281971454621,
|
|
"num_tokens": 18615165.0,
|
|
"step": 8425
|
|
},
|
|
{
|
|
"entropy": 5.105244731903076,
|
|
"epoch": 8.498234997478567,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 7.998581170262412e-05,
|
|
"loss": 4.3355,
|
|
"mean_token_accuracy": 0.24195968806743623,
|
|
"num_tokens": 18625853.0,
|
|
"step": 8430
|
|
},
|
|
{
|
|
"entropy": 5.147421073913574,
|
|
"epoch": 8.503277861825516,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.978826667993141e-05,
|
|
"loss": 4.3958,
|
|
"mean_token_accuracy": 0.23002272099256516,
|
|
"num_tokens": 18635285.0,
|
|
"step": 8435
|
|
},
|
|
{
|
|
"entropy": 5.142104959487915,
|
|
"epoch": 8.508320726172466,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 7.95913283798091e-05,
|
|
"loss": 4.4041,
|
|
"mean_token_accuracy": 0.22855813801288605,
|
|
"num_tokens": 18646983.0,
|
|
"step": 8440
|
|
},
|
|
{
|
|
"entropy": 5.189692687988281,
|
|
"epoch": 8.513363590519415,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 7.93949974143458e-05,
|
|
"loss": 4.4052,
|
|
"mean_token_accuracy": 0.226900215446949,
|
|
"num_tokens": 18657118.0,
|
|
"step": 8445
|
|
},
|
|
{
|
|
"entropy": 5.105091667175293,
|
|
"epoch": 8.518406454866364,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 7.919927439374278e-05,
|
|
"loss": 4.3885,
|
|
"mean_token_accuracy": 0.23132576942443847,
|
|
"num_tokens": 18667942.0,
|
|
"step": 8450
|
|
},
|
|
{
|
|
"entropy": 5.138662385940552,
|
|
"epoch": 8.523449319213313,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 7.900415992631162e-05,
|
|
"loss": 4.4416,
|
|
"mean_token_accuracy": 0.2182560846209526,
|
|
"num_tokens": 18678553.0,
|
|
"step": 8455
|
|
},
|
|
{
|
|
"entropy": 5.167260694503784,
|
|
"epoch": 8.528492183560262,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7.880965461847261e-05,
|
|
"loss": 4.4172,
|
|
"mean_token_accuracy": 0.2214689001441002,
|
|
"num_tokens": 18690457.0,
|
|
"step": 8460
|
|
},
|
|
{
|
|
"entropy": 5.16165337562561,
|
|
"epoch": 8.533535047907211,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 7.86157590747528e-05,
|
|
"loss": 4.3997,
|
|
"mean_token_accuracy": 0.22962938994169235,
|
|
"num_tokens": 18703215.0,
|
|
"step": 8465
|
|
},
|
|
{
|
|
"entropy": 5.125035524368286,
|
|
"epoch": 8.53857791225416,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 7.8422473897784e-05,
|
|
"loss": 4.3561,
|
|
"mean_token_accuracy": 0.2325684145092964,
|
|
"num_tokens": 18713180.0,
|
|
"step": 8470
|
|
},
|
|
{
|
|
"entropy": 5.149163722991943,
|
|
"epoch": 8.54362077660111,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 7.822979968830089e-05,
|
|
"loss": 4.4419,
|
|
"mean_token_accuracy": 0.22423619776964188,
|
|
"num_tokens": 18725928.0,
|
|
"step": 8475
|
|
},
|
|
{
|
|
"entropy": 5.179641246795654,
|
|
"epoch": 8.548663640948059,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 7.803773704513953e-05,
|
|
"loss": 4.4732,
|
|
"mean_token_accuracy": 0.21907952427864075,
|
|
"num_tokens": 18737321.0,
|
|
"step": 8480
|
|
},
|
|
{
|
|
"entropy": 5.199834203720092,
|
|
"epoch": 8.553706505295008,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 7.784628656523488e-05,
|
|
"loss": 4.4721,
|
|
"mean_token_accuracy": 0.22604528665542603,
|
|
"num_tokens": 18748200.0,
|
|
"step": 8485
|
|
},
|
|
{
|
|
"entropy": 5.181067943572998,
|
|
"epoch": 8.558749369641957,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 7.76554488436195e-05,
|
|
"loss": 4.4415,
|
|
"mean_token_accuracy": 0.22350989431142806,
|
|
"num_tokens": 18759281.0,
|
|
"step": 8490
|
|
},
|
|
{
|
|
"entropy": 5.115938711166382,
|
|
"epoch": 8.563792233988906,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7.746522447342148e-05,
|
|
"loss": 4.384,
|
|
"mean_token_accuracy": 0.22448837608098984,
|
|
"num_tokens": 18771089.0,
|
|
"step": 8495
|
|
},
|
|
{
|
|
"entropy": 5.134475231170654,
|
|
"epoch": 8.568835098335855,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 7.727561404586255e-05,
|
|
"loss": 4.4097,
|
|
"mean_token_accuracy": 0.2263183295726776,
|
|
"num_tokens": 18781619.0,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"epoch": 8.568835098335855,
|
|
"eval_entropy": 5.093182471568954,
|
|
"eval_loss": 5.6973347663879395,
|
|
"eval_mean_token_accuracy": 0.15561470290783405,
|
|
"eval_num_tokens": 18781619.0,
|
|
"eval_runtime": 2.6507,
|
|
"eval_samples_per_second": 1183.854,
|
|
"eval_steps_per_second": 148.265,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"entropy": 5.160215663909912,
|
|
"epoch": 8.573877962682804,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 7.708661815025619e-05,
|
|
"loss": 4.4487,
|
|
"mean_token_accuracy": 0.2267065703868866,
|
|
"num_tokens": 18793828.0,
|
|
"step": 8505
|
|
},
|
|
{
|
|
"entropy": 5.126705121994019,
|
|
"epoch": 8.578920827029753,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 7.689823737400617e-05,
|
|
"loss": 4.4177,
|
|
"mean_token_accuracy": 0.2283346176147461,
|
|
"num_tokens": 18804134.0,
|
|
"step": 8510
|
|
},
|
|
{
|
|
"entropy": 5.101263618469238,
|
|
"epoch": 8.583963691376702,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 7.671047230260412e-05,
|
|
"loss": 4.3397,
|
|
"mean_token_accuracy": 0.23861866742372512,
|
|
"num_tokens": 18815576.0,
|
|
"step": 8515
|
|
},
|
|
{
|
|
"entropy": 5.135848474502564,
|
|
"epoch": 8.589006555723651,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.652332351962836e-05,
|
|
"loss": 4.409,
|
|
"mean_token_accuracy": 0.22318531572818756,
|
|
"num_tokens": 18825820.0,
|
|
"step": 8520
|
|
},
|
|
{
|
|
"entropy": 5.171012210845947,
|
|
"epoch": 8.5940494200706,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.633679160674144e-05,
|
|
"loss": 4.4495,
|
|
"mean_token_accuracy": 0.22282206416130065,
|
|
"num_tokens": 18837597.0,
|
|
"step": 8525
|
|
},
|
|
{
|
|
"entropy": 5.116401481628418,
|
|
"epoch": 8.59909228441755,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 7.615087714368902e-05,
|
|
"loss": 4.4102,
|
|
"mean_token_accuracy": 0.22676160484552382,
|
|
"num_tokens": 18848988.0,
|
|
"step": 8530
|
|
},
|
|
{
|
|
"entropy": 5.128191900253296,
|
|
"epoch": 8.604135148764499,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 7.59655807082975e-05,
|
|
"loss": 4.4684,
|
|
"mean_token_accuracy": 0.22179030776023864,
|
|
"num_tokens": 18860274.0,
|
|
"step": 8535
|
|
},
|
|
{
|
|
"entropy": 5.084139633178711,
|
|
"epoch": 8.609178013111448,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 7.578090287647242e-05,
|
|
"loss": 4.372,
|
|
"mean_token_accuracy": 0.22947443127632142,
|
|
"num_tokens": 18871272.0,
|
|
"step": 8540
|
|
},
|
|
{
|
|
"entropy": 5.1470703125,
|
|
"epoch": 8.614220877458397,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7.559684422219683e-05,
|
|
"loss": 4.3921,
|
|
"mean_token_accuracy": 0.231125371158123,
|
|
"num_tokens": 18882675.0,
|
|
"step": 8545
|
|
},
|
|
{
|
|
"entropy": 5.104860448837281,
|
|
"epoch": 8.619263741805346,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.541340531752916e-05,
|
|
"loss": 4.4069,
|
|
"mean_token_accuracy": 0.22715360224246978,
|
|
"num_tokens": 18893767.0,
|
|
"step": 8550
|
|
},
|
|
{
|
|
"entropy": 5.149162673950196,
|
|
"epoch": 8.624306606152295,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 7.523058673260183e-05,
|
|
"loss": 4.4351,
|
|
"mean_token_accuracy": 0.22500626146793365,
|
|
"num_tokens": 18904365.0,
|
|
"step": 8555
|
|
},
|
|
{
|
|
"entropy": 5.118316173553467,
|
|
"epoch": 8.629349470499244,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.50483890356192e-05,
|
|
"loss": 4.4099,
|
|
"mean_token_accuracy": 0.2263246163725853,
|
|
"num_tokens": 18914364.0,
|
|
"step": 8560
|
|
},
|
|
{
|
|
"entropy": 5.121327447891235,
|
|
"epoch": 8.634392334846194,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7.486681279285598e-05,
|
|
"loss": 4.4548,
|
|
"mean_token_accuracy": 0.22448006719350816,
|
|
"num_tokens": 18926073.0,
|
|
"step": 8565
|
|
},
|
|
{
|
|
"entropy": 5.105640983581543,
|
|
"epoch": 8.63943519919314,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.468585856865522e-05,
|
|
"loss": 4.3623,
|
|
"mean_token_accuracy": 0.23314609676599501,
|
|
"num_tokens": 18936174.0,
|
|
"step": 8570
|
|
},
|
|
{
|
|
"entropy": 5.11095814704895,
|
|
"epoch": 8.64447806354009,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.45055269254269e-05,
|
|
"loss": 4.3464,
|
|
"mean_token_accuracy": 0.23871779292821885,
|
|
"num_tokens": 18946875.0,
|
|
"step": 8575
|
|
},
|
|
{
|
|
"entropy": 5.118292379379272,
|
|
"epoch": 8.649520927887039,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 7.432581842364585e-05,
|
|
"loss": 4.3607,
|
|
"mean_token_accuracy": 0.23290147632360458,
|
|
"num_tokens": 18958744.0,
|
|
"step": 8580
|
|
},
|
|
{
|
|
"entropy": 5.10329852104187,
|
|
"epoch": 8.654563792233988,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 7.414673362185031e-05,
|
|
"loss": 4.3794,
|
|
"mean_token_accuracy": 0.22310455441474913,
|
|
"num_tokens": 18970107.0,
|
|
"step": 8585
|
|
},
|
|
{
|
|
"entropy": 5.203870105743408,
|
|
"epoch": 8.659606656580937,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.396827307663988e-05,
|
|
"loss": 4.4813,
|
|
"mean_token_accuracy": 0.22412588447332382,
|
|
"num_tokens": 18980736.0,
|
|
"step": 8590
|
|
},
|
|
{
|
|
"entropy": 5.172149038314819,
|
|
"epoch": 8.664649520927886,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.379043734267421e-05,
|
|
"loss": 4.4364,
|
|
"mean_token_accuracy": 0.21972141414880753,
|
|
"num_tokens": 18991029.0,
|
|
"step": 8595
|
|
},
|
|
{
|
|
"entropy": 5.14033260345459,
|
|
"epoch": 8.669692385274836,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 7.36132269726707e-05,
|
|
"loss": 4.4176,
|
|
"mean_token_accuracy": 0.22649382501840593,
|
|
"num_tokens": 19000872.0,
|
|
"step": 8600
|
|
},
|
|
{
|
|
"entropy": 5.16853518486023,
|
|
"epoch": 8.674735249621785,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.343664251740342e-05,
|
|
"loss": 4.4527,
|
|
"mean_token_accuracy": 0.22074526995420457,
|
|
"num_tokens": 19011954.0,
|
|
"step": 8605
|
|
},
|
|
{
|
|
"entropy": 5.16057596206665,
|
|
"epoch": 8.679778113968734,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.326068452570078e-05,
|
|
"loss": 4.3902,
|
|
"mean_token_accuracy": 0.22846103459596634,
|
|
"num_tokens": 19024641.0,
|
|
"step": 8610
|
|
},
|
|
{
|
|
"entropy": 5.182190656661987,
|
|
"epoch": 8.684820978315683,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 7.30853535444444e-05,
|
|
"loss": 4.441,
|
|
"mean_token_accuracy": 0.22321225255727767,
|
|
"num_tokens": 19035639.0,
|
|
"step": 8615
|
|
},
|
|
{
|
|
"entropy": 5.122605562210083,
|
|
"epoch": 8.689863842662632,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 7.29106501185669e-05,
|
|
"loss": 4.3697,
|
|
"mean_token_accuracy": 0.22855599522590636,
|
|
"num_tokens": 19046020.0,
|
|
"step": 8620
|
|
},
|
|
{
|
|
"entropy": 5.172622728347778,
|
|
"epoch": 8.694906707009581,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 7.273657479105068e-05,
|
|
"loss": 4.4238,
|
|
"mean_token_accuracy": 0.2271439641714096,
|
|
"num_tokens": 19057012.0,
|
|
"step": 8625
|
|
},
|
|
{
|
|
"entropy": 5.190731334686279,
|
|
"epoch": 8.69994957135653,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.256312810292576e-05,
|
|
"loss": 4.4247,
|
|
"mean_token_accuracy": 0.2292873203754425,
|
|
"num_tokens": 19066225.0,
|
|
"step": 8630
|
|
},
|
|
{
|
|
"entropy": 5.164949607849121,
|
|
"epoch": 8.70499243570348,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 7.23903105932685e-05,
|
|
"loss": 4.4449,
|
|
"mean_token_accuracy": 0.2246780201792717,
|
|
"num_tokens": 19076457.0,
|
|
"step": 8635
|
|
},
|
|
{
|
|
"entropy": 5.140701103210449,
|
|
"epoch": 8.710035300050428,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 7.221812279919978e-05,
|
|
"loss": 4.3496,
|
|
"mean_token_accuracy": 0.22969255596399307,
|
|
"num_tokens": 19087572.0,
|
|
"step": 8640
|
|
},
|
|
{
|
|
"entropy": 5.111488008499146,
|
|
"epoch": 8.715078164397378,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 7.20465652558831e-05,
|
|
"loss": 4.3436,
|
|
"mean_token_accuracy": 0.2323598027229309,
|
|
"num_tokens": 19099120.0,
|
|
"step": 8645
|
|
},
|
|
{
|
|
"entropy": 5.221729421615601,
|
|
"epoch": 8.720121028744327,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.187563849652332e-05,
|
|
"loss": 4.5199,
|
|
"mean_token_accuracy": 0.21651394069194793,
|
|
"num_tokens": 19110540.0,
|
|
"step": 8650
|
|
},
|
|
{
|
|
"entropy": 5.149336433410644,
|
|
"epoch": 8.725163893091276,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.170534305236479e-05,
|
|
"loss": 4.3754,
|
|
"mean_token_accuracy": 0.2344029352068901,
|
|
"num_tokens": 19121106.0,
|
|
"step": 8655
|
|
},
|
|
{
|
|
"entropy": 5.128800582885742,
|
|
"epoch": 8.730206757438225,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 7.153567945268963e-05,
|
|
"loss": 4.3654,
|
|
"mean_token_accuracy": 0.2321561485528946,
|
|
"num_tokens": 19132175.0,
|
|
"step": 8660
|
|
},
|
|
{
|
|
"entropy": 5.13642373085022,
|
|
"epoch": 8.735249621785174,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.136664822481624e-05,
|
|
"loss": 4.4388,
|
|
"mean_token_accuracy": 0.22429257929325103,
|
|
"num_tokens": 19144180.0,
|
|
"step": 8665
|
|
},
|
|
{
|
|
"entropy": 5.1021631240844725,
|
|
"epoch": 8.740292486132123,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7.119824989409761e-05,
|
|
"loss": 4.331,
|
|
"mean_token_accuracy": 0.2332303136587143,
|
|
"num_tokens": 19154763.0,
|
|
"step": 8670
|
|
},
|
|
{
|
|
"entropy": 5.128591632843017,
|
|
"epoch": 8.745335350479072,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 7.103048498391954e-05,
|
|
"loss": 4.4468,
|
|
"mean_token_accuracy": 0.22137288600206376,
|
|
"num_tokens": 19166551.0,
|
|
"step": 8675
|
|
},
|
|
{
|
|
"entropy": 5.091091966629028,
|
|
"epoch": 8.750378214826021,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 7.086335401569935e-05,
|
|
"loss": 4.3544,
|
|
"mean_token_accuracy": 0.22827789634466172,
|
|
"num_tokens": 19177816.0,
|
|
"step": 8680
|
|
},
|
|
{
|
|
"entropy": 5.113712358474731,
|
|
"epoch": 8.75542107917297,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 7.06968575088838e-05,
|
|
"loss": 4.3407,
|
|
"mean_token_accuracy": 0.23288343250751495,
|
|
"num_tokens": 19187305.0,
|
|
"step": 8685
|
|
},
|
|
{
|
|
"entropy": 5.150473499298096,
|
|
"epoch": 8.76046394351992,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.053099598094804e-05,
|
|
"loss": 4.3923,
|
|
"mean_token_accuracy": 0.22407064437866211,
|
|
"num_tokens": 19196872.0,
|
|
"step": 8690
|
|
},
|
|
{
|
|
"entropy": 5.147425746917724,
|
|
"epoch": 8.765506807866869,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.036576994739343e-05,
|
|
"loss": 4.4461,
|
|
"mean_token_accuracy": 0.22298524975776673,
|
|
"num_tokens": 19208046.0,
|
|
"step": 8695
|
|
},
|
|
{
|
|
"entropy": 5.196120405197144,
|
|
"epoch": 8.770549672213818,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 7.020117992174638e-05,
|
|
"loss": 4.5195,
|
|
"mean_token_accuracy": 0.2130984589457512,
|
|
"num_tokens": 19219102.0,
|
|
"step": 8700
|
|
},
|
|
{
|
|
"entropy": 5.150086355209351,
|
|
"epoch": 8.775592536560767,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 7.003722641555635e-05,
|
|
"loss": 4.3763,
|
|
"mean_token_accuracy": 0.22778175622224808,
|
|
"num_tokens": 19229709.0,
|
|
"step": 8705
|
|
},
|
|
{
|
|
"entropy": 5.167554426193237,
|
|
"epoch": 8.780635400907716,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 6.987390993839476e-05,
|
|
"loss": 4.3727,
|
|
"mean_token_accuracy": 0.22750797122716904,
|
|
"num_tokens": 19240361.0,
|
|
"step": 8710
|
|
},
|
|
{
|
|
"entropy": 5.143374300003051,
|
|
"epoch": 8.785678265254665,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 6.971123099785281e-05,
|
|
"loss": 4.4315,
|
|
"mean_token_accuracy": 0.23027579337358475,
|
|
"num_tokens": 19252470.0,
|
|
"step": 8715
|
|
},
|
|
{
|
|
"entropy": 5.142523717880249,
|
|
"epoch": 8.790721129601614,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 6.954919009954061e-05,
|
|
"loss": 4.4175,
|
|
"mean_token_accuracy": 0.22388386577367783,
|
|
"num_tokens": 19264529.0,
|
|
"step": 8720
|
|
},
|
|
{
|
|
"entropy": 5.127300119400024,
|
|
"epoch": 8.795763993948563,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 6.938778774708488e-05,
|
|
"loss": 4.3154,
|
|
"mean_token_accuracy": 0.23830656260251998,
|
|
"num_tokens": 19275383.0,
|
|
"step": 8725
|
|
},
|
|
{
|
|
"entropy": 5.157727956771851,
|
|
"epoch": 8.800806858295513,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 6.922702444212792e-05,
|
|
"loss": 4.4217,
|
|
"mean_token_accuracy": 0.22509915083646775,
|
|
"num_tokens": 19286571.0,
|
|
"step": 8730
|
|
},
|
|
{
|
|
"entropy": 5.114703941345215,
|
|
"epoch": 8.805849722642462,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 6.906690068432568e-05,
|
|
"loss": 4.4526,
|
|
"mean_token_accuracy": 0.22155337333679198,
|
|
"num_tokens": 19297186.0,
|
|
"step": 8735
|
|
},
|
|
{
|
|
"entropy": 5.196652889251709,
|
|
"epoch": 8.81089258698941,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 6.89074169713466e-05,
|
|
"loss": 4.4969,
|
|
"mean_token_accuracy": 0.22668879479169846,
|
|
"num_tokens": 19308554.0,
|
|
"step": 8740
|
|
},
|
|
{
|
|
"entropy": 5.18836727142334,
|
|
"epoch": 8.815935451336358,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 6.87485737988697e-05,
|
|
"loss": 4.485,
|
|
"mean_token_accuracy": 0.22287349849939347,
|
|
"num_tokens": 19319095.0,
|
|
"step": 8745
|
|
},
|
|
{
|
|
"entropy": 5.152085018157959,
|
|
"epoch": 8.820978315683309,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 6.859037166058323e-05,
|
|
"loss": 4.3721,
|
|
"mean_token_accuracy": 0.22942841649055482,
|
|
"num_tokens": 19331235.0,
|
|
"step": 8750
|
|
},
|
|
{
|
|
"entropy": 5.1631255626678465,
|
|
"epoch": 8.826021180030256,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.843281104818313e-05,
|
|
"loss": 4.405,
|
|
"mean_token_accuracy": 0.22446652501821518,
|
|
"num_tokens": 19341319.0,
|
|
"step": 8755
|
|
},
|
|
{
|
|
"entropy": 5.140712642669678,
|
|
"epoch": 8.831064044377205,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 6.827589245137135e-05,
|
|
"loss": 4.3627,
|
|
"mean_token_accuracy": 0.23000483959913254,
|
|
"num_tokens": 19352097.0,
|
|
"step": 8760
|
|
},
|
|
{
|
|
"entropy": 5.107294082641602,
|
|
"epoch": 8.836106908724155,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 6.811961635785456e-05,
|
|
"loss": 4.3788,
|
|
"mean_token_accuracy": 0.22919373363256454,
|
|
"num_tokens": 19363560.0,
|
|
"step": 8765
|
|
},
|
|
{
|
|
"entropy": 5.127737808227539,
|
|
"epoch": 8.841149773071104,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 6.796398325334238e-05,
|
|
"loss": 4.3409,
|
|
"mean_token_accuracy": 0.2344162493944168,
|
|
"num_tokens": 19374783.0,
|
|
"step": 8770
|
|
},
|
|
{
|
|
"entropy": 5.124257564544678,
|
|
"epoch": 8.846192637418053,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 6.78089936215462e-05,
|
|
"loss": 4.3663,
|
|
"mean_token_accuracy": 0.23397525101900102,
|
|
"num_tokens": 19386025.0,
|
|
"step": 8775
|
|
},
|
|
{
|
|
"entropy": 5.12721734046936,
|
|
"epoch": 8.851235501765002,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 6.765464794417723e-05,
|
|
"loss": 4.4499,
|
|
"mean_token_accuracy": 0.21956741213798522,
|
|
"num_tokens": 19398524.0,
|
|
"step": 8780
|
|
},
|
|
{
|
|
"entropy": 5.178207683563232,
|
|
"epoch": 8.856278366111951,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 6.750094670094558e-05,
|
|
"loss": 4.4296,
|
|
"mean_token_accuracy": 0.2256477504968643,
|
|
"num_tokens": 19409262.0,
|
|
"step": 8785
|
|
},
|
|
{
|
|
"entropy": 5.128272438049317,
|
|
"epoch": 8.8613212304589,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 6.734789036955813e-05,
|
|
"loss": 4.3705,
|
|
"mean_token_accuracy": 0.23649449944496154,
|
|
"num_tokens": 19421559.0,
|
|
"step": 8790
|
|
},
|
|
{
|
|
"entropy": 5.1493181705474855,
|
|
"epoch": 8.86636409480585,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.719547942571755e-05,
|
|
"loss": 4.4091,
|
|
"mean_token_accuracy": 0.22492718547582627,
|
|
"num_tokens": 19433328.0,
|
|
"step": 8795
|
|
},
|
|
{
|
|
"entropy": 5.135502862930298,
|
|
"epoch": 8.871406959152798,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 6.704371434312048e-05,
|
|
"loss": 4.4458,
|
|
"mean_token_accuracy": 0.2226995572447777,
|
|
"num_tokens": 19444258.0,
|
|
"step": 8800
|
|
},
|
|
{
|
|
"entropy": 5.100721406936645,
|
|
"epoch": 8.876449823499748,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 6.689259559345641e-05,
|
|
"loss": 4.3263,
|
|
"mean_token_accuracy": 0.23668635487556458,
|
|
"num_tokens": 19455084.0,
|
|
"step": 8805
|
|
},
|
|
{
|
|
"entropy": 5.177768182754517,
|
|
"epoch": 8.881492687846697,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.674212364640577e-05,
|
|
"loss": 4.4358,
|
|
"mean_token_accuracy": 0.2208070158958435,
|
|
"num_tokens": 19466064.0,
|
|
"step": 8810
|
|
},
|
|
{
|
|
"entropy": 5.1354881763458256,
|
|
"epoch": 8.886535552193646,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 6.659229896963899e-05,
|
|
"loss": 4.3989,
|
|
"mean_token_accuracy": 0.23140463083982468,
|
|
"num_tokens": 19477146.0,
|
|
"step": 8815
|
|
},
|
|
{
|
|
"entropy": 5.214872074127197,
|
|
"epoch": 8.891578416540595,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 6.644312202881452e-05,
|
|
"loss": 4.4897,
|
|
"mean_token_accuracy": 0.22057319432497025,
|
|
"num_tokens": 19487859.0,
|
|
"step": 8820
|
|
},
|
|
{
|
|
"entropy": 5.125205230712891,
|
|
"epoch": 8.896621280887544,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 6.629459328757781e-05,
|
|
"loss": 4.3694,
|
|
"mean_token_accuracy": 0.2336997628211975,
|
|
"num_tokens": 19498573.0,
|
|
"step": 8825
|
|
},
|
|
{
|
|
"entropy": 5.139772272109985,
|
|
"epoch": 8.901664145234493,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.614671320755953e-05,
|
|
"loss": 4.3778,
|
|
"mean_token_accuracy": 0.22906025052070617,
|
|
"num_tokens": 19509592.0,
|
|
"step": 8830
|
|
},
|
|
{
|
|
"entropy": 5.186947631835937,
|
|
"epoch": 8.906707009581442,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.599948224837446e-05,
|
|
"loss": 4.4594,
|
|
"mean_token_accuracy": 0.22208300977945328,
|
|
"num_tokens": 19520621.0,
|
|
"step": 8835
|
|
},
|
|
{
|
|
"entropy": 5.163477373123169,
|
|
"epoch": 8.911749873928391,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 6.585290086761972e-05,
|
|
"loss": 4.4133,
|
|
"mean_token_accuracy": 0.23000758290290832,
|
|
"num_tokens": 19532678.0,
|
|
"step": 8840
|
|
},
|
|
{
|
|
"entropy": 5.140813398361206,
|
|
"epoch": 8.91679273827534,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 6.570696952087368e-05,
|
|
"loss": 4.3519,
|
|
"mean_token_accuracy": 0.23263231217861174,
|
|
"num_tokens": 19541840.0,
|
|
"step": 8845
|
|
},
|
|
{
|
|
"entropy": 5.097993564605713,
|
|
"epoch": 8.92183560262229,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 6.556168866169435e-05,
|
|
"loss": 4.3428,
|
|
"mean_token_accuracy": 0.23207785934209824,
|
|
"num_tokens": 19553198.0,
|
|
"step": 8850
|
|
},
|
|
{
|
|
"entropy": 5.126378107070923,
|
|
"epoch": 8.926878466969239,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 6.541705874161795e-05,
|
|
"loss": 4.3131,
|
|
"mean_token_accuracy": 0.23537041544914244,
|
|
"num_tokens": 19562996.0,
|
|
"step": 8855
|
|
},
|
|
{
|
|
"entropy": 5.133792734146118,
|
|
"epoch": 8.931921331316188,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 6.527308021015771e-05,
|
|
"loss": 4.4136,
|
|
"mean_token_accuracy": 0.23184903264045714,
|
|
"num_tokens": 19573486.0,
|
|
"step": 8860
|
|
},
|
|
{
|
|
"entropy": 5.139494562149048,
|
|
"epoch": 8.936964195663137,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 6.512975351480207e-05,
|
|
"loss": 4.5017,
|
|
"mean_token_accuracy": 0.21672467589378358,
|
|
"num_tokens": 19584026.0,
|
|
"step": 8865
|
|
},
|
|
{
|
|
"entropy": 5.106271553039551,
|
|
"epoch": 8.942007060010086,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.498707910101383e-05,
|
|
"loss": 4.3484,
|
|
"mean_token_accuracy": 0.22965519577264787,
|
|
"num_tokens": 19594766.0,
|
|
"step": 8870
|
|
},
|
|
{
|
|
"entropy": 5.149125003814698,
|
|
"epoch": 8.947049924357035,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.484505741222824e-05,
|
|
"loss": 4.3582,
|
|
"mean_token_accuracy": 0.24234689623117447,
|
|
"num_tokens": 19604069.0,
|
|
"step": 8875
|
|
},
|
|
{
|
|
"entropy": 5.100041723251342,
|
|
"epoch": 8.952092788703984,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.470368888985215e-05,
|
|
"loss": 4.4347,
|
|
"mean_token_accuracy": 0.21716218739748,
|
|
"num_tokens": 19614830.0,
|
|
"step": 8880
|
|
},
|
|
{
|
|
"entropy": 5.196165418624878,
|
|
"epoch": 8.957135653050933,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 6.456297397326202e-05,
|
|
"loss": 4.5022,
|
|
"mean_token_accuracy": 0.21616822332143784,
|
|
"num_tokens": 19627803.0,
|
|
"step": 8885
|
|
},
|
|
{
|
|
"entropy": 5.143807744979858,
|
|
"epoch": 8.962178517397883,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 6.442291309980321e-05,
|
|
"loss": 4.4441,
|
|
"mean_token_accuracy": 0.2175200790166855,
|
|
"num_tokens": 19639391.0,
|
|
"step": 8890
|
|
},
|
|
{
|
|
"entropy": 5.138750314712524,
|
|
"epoch": 8.967221381744832,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.4283506704788e-05,
|
|
"loss": 4.3606,
|
|
"mean_token_accuracy": 0.23137067556381224,
|
|
"num_tokens": 19649582.0,
|
|
"step": 8895
|
|
},
|
|
{
|
|
"entropy": 5.238546228408813,
|
|
"epoch": 8.97226424609178,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 6.414475522149481e-05,
|
|
"loss": 4.5103,
|
|
"mean_token_accuracy": 0.21534627676010132,
|
|
"num_tokens": 19661900.0,
|
|
"step": 8900
|
|
},
|
|
{
|
|
"entropy": 5.186837148666382,
|
|
"epoch": 8.97730711043873,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 6.400665908116631e-05,
|
|
"loss": 4.4568,
|
|
"mean_token_accuracy": 0.22098565250635147,
|
|
"num_tokens": 19672980.0,
|
|
"step": 8905
|
|
},
|
|
{
|
|
"entropy": 5.1692688941955565,
|
|
"epoch": 8.982349974785679,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 6.386921871300865e-05,
|
|
"loss": 4.4293,
|
|
"mean_token_accuracy": 0.22530302703380584,
|
|
"num_tokens": 19684242.0,
|
|
"step": 8910
|
|
},
|
|
{
|
|
"entropy": 5.153019809722901,
|
|
"epoch": 8.987392839132628,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 6.373243454418952e-05,
|
|
"loss": 4.3865,
|
|
"mean_token_accuracy": 0.2225500911474228,
|
|
"num_tokens": 19696043.0,
|
|
"step": 8915
|
|
},
|
|
{
|
|
"entropy": 5.08813796043396,
|
|
"epoch": 8.992435703479577,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 6.35963069998374e-05,
|
|
"loss": 4.2811,
|
|
"mean_token_accuracy": 0.24174141585826875,
|
|
"num_tokens": 19706963.0,
|
|
"step": 8920
|
|
},
|
|
{
|
|
"entropy": 5.143728303909302,
|
|
"epoch": 8.997478567826526,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 6.346083650303975e-05,
|
|
"loss": 4.3931,
|
|
"mean_token_accuracy": 0.22891862839460372,
|
|
"num_tokens": 19720544.0,
|
|
"step": 8925
|
|
},
|
|
{
|
|
"entropy": 5.152339511447483,
|
|
"epoch": 9.00201714573878,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 6.332602347484202e-05,
|
|
"loss": 4.2784,
|
|
"mean_token_accuracy": 0.23232066134611765,
|
|
"num_tokens": 19728776.0,
|
|
"step": 8930
|
|
},
|
|
{
|
|
"entropy": 5.12672529220581,
|
|
"epoch": 9.007060010085729,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.31918683342462e-05,
|
|
"loss": 4.3683,
|
|
"mean_token_accuracy": 0.23191899210214614,
|
|
"num_tokens": 19739465.0,
|
|
"step": 8935
|
|
},
|
|
{
|
|
"entropy": 5.156985425949097,
|
|
"epoch": 9.012102874432678,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 6.305837149820959e-05,
|
|
"loss": 4.3891,
|
|
"mean_token_accuracy": 0.22936780303716658,
|
|
"num_tokens": 19751000.0,
|
|
"step": 8940
|
|
},
|
|
{
|
|
"entropy": 5.119020795822143,
|
|
"epoch": 9.017145738779627,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.29255333816434e-05,
|
|
"loss": 4.3441,
|
|
"mean_token_accuracy": 0.23126667141914367,
|
|
"num_tokens": 19762980.0,
|
|
"step": 8945
|
|
},
|
|
{
|
|
"entropy": 5.151031494140625,
|
|
"epoch": 9.022188603126576,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 6.279335439741164e-05,
|
|
"loss": 4.4222,
|
|
"mean_token_accuracy": 0.22738688737154006,
|
|
"num_tokens": 19774725.0,
|
|
"step": 8950
|
|
},
|
|
{
|
|
"entropy": 5.10569658279419,
|
|
"epoch": 9.027231467473525,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.266183495632954e-05,
|
|
"loss": 4.3252,
|
|
"mean_token_accuracy": 0.22828114479780198,
|
|
"num_tokens": 19784970.0,
|
|
"step": 8955
|
|
},
|
|
{
|
|
"entropy": 5.138676023483276,
|
|
"epoch": 9.032274331820474,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 6.253097546716264e-05,
|
|
"loss": 4.3973,
|
|
"mean_token_accuracy": 0.2339349567890167,
|
|
"num_tokens": 19797148.0,
|
|
"step": 8960
|
|
},
|
|
{
|
|
"entropy": 5.159294271469117,
|
|
"epoch": 9.037317196167423,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.240077633662519e-05,
|
|
"loss": 4.33,
|
|
"mean_token_accuracy": 0.22733146995306014,
|
|
"num_tokens": 19806256.0,
|
|
"step": 8965
|
|
},
|
|
{
|
|
"entropy": 5.141302299499512,
|
|
"epoch": 9.042360060514373,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.227123796937905e-05,
|
|
"loss": 4.3567,
|
|
"mean_token_accuracy": 0.2322322264313698,
|
|
"num_tokens": 19816383.0,
|
|
"step": 8970
|
|
},
|
|
{
|
|
"entropy": 5.133065271377563,
|
|
"epoch": 9.047402924861322,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 6.214236076803264e-05,
|
|
"loss": 4.3219,
|
|
"mean_token_accuracy": 0.23727035373449326,
|
|
"num_tokens": 19826388.0,
|
|
"step": 8975
|
|
},
|
|
{
|
|
"entropy": 5.088606977462769,
|
|
"epoch": 9.05244578920827,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 6.201414513313907e-05,
|
|
"loss": 4.333,
|
|
"mean_token_accuracy": 0.23602795451879502,
|
|
"num_tokens": 19836894.0,
|
|
"step": 8980
|
|
},
|
|
{
|
|
"entropy": 5.114261865615845,
|
|
"epoch": 9.05748865355522,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.188659146319562e-05,
|
|
"loss": 4.3443,
|
|
"mean_token_accuracy": 0.23264195024967194,
|
|
"num_tokens": 19848311.0,
|
|
"step": 8985
|
|
},
|
|
{
|
|
"entropy": 5.148709869384765,
|
|
"epoch": 9.062531517902169,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 6.175970015464196e-05,
|
|
"loss": 4.3576,
|
|
"mean_token_accuracy": 0.22448411881923674,
|
|
"num_tokens": 19859555.0,
|
|
"step": 8990
|
|
},
|
|
{
|
|
"entropy": 5.153161859512329,
|
|
"epoch": 9.067574382249118,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 6.163347160185929e-05,
|
|
"loss": 4.3411,
|
|
"mean_token_accuracy": 0.23308941572904587,
|
|
"num_tokens": 19870557.0,
|
|
"step": 8995
|
|
},
|
|
{
|
|
"entropy": 5.137364625930786,
|
|
"epoch": 9.072617246596067,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.150790619716873e-05,
|
|
"loss": 4.416,
|
|
"mean_token_accuracy": 0.22408677339553834,
|
|
"num_tokens": 19881508.0,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"epoch": 9.072617246596067,
|
|
"eval_entropy": 5.082711037788682,
|
|
"eval_loss": 5.697414875030518,
|
|
"eval_mean_token_accuracy": 0.15602438959456583,
|
|
"eval_num_tokens": 19881508.0,
|
|
"eval_runtime": 2.6564,
|
|
"eval_samples_per_second": 1181.279,
|
|
"eval_steps_per_second": 147.942,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"entropy": 5.149577951431274,
|
|
"epoch": 9.077660110943016,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 6.138300433083061e-05,
|
|
"loss": 4.3787,
|
|
"mean_token_accuracy": 0.23333051800727844,
|
|
"num_tokens": 19890459.0,
|
|
"step": 9005
|
|
},
|
|
{
|
|
"entropy": 5.092114877700806,
|
|
"epoch": 9.082702975289966,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 6.12587663910427e-05,
|
|
"loss": 4.3566,
|
|
"mean_token_accuracy": 0.23137201368808746,
|
|
"num_tokens": 19901384.0,
|
|
"step": 9010
|
|
},
|
|
{
|
|
"entropy": 5.032909488677978,
|
|
"epoch": 9.087745839636915,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.113519276393942e-05,
|
|
"loss": 4.2531,
|
|
"mean_token_accuracy": 0.2448873072862625,
|
|
"num_tokens": 19911398.0,
|
|
"step": 9015
|
|
},
|
|
{
|
|
"entropy": 5.185895395278931,
|
|
"epoch": 9.092788703983862,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.1012283833590465e-05,
|
|
"loss": 4.3731,
|
|
"mean_token_accuracy": 0.23187076151371003,
|
|
"num_tokens": 19923102.0,
|
|
"step": 9020
|
|
},
|
|
{
|
|
"entropy": 5.138507604598999,
|
|
"epoch": 9.097831568330811,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 6.089003998199961e-05,
|
|
"loss": 4.3627,
|
|
"mean_token_accuracy": 0.23140767216682434,
|
|
"num_tokens": 19933641.0,
|
|
"step": 9025
|
|
},
|
|
{
|
|
"entropy": 5.1768065929412845,
|
|
"epoch": 9.10287443267776,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 6.076846158910357e-05,
|
|
"loss": 4.4434,
|
|
"mean_token_accuracy": 0.22096142768859864,
|
|
"num_tokens": 19944993.0,
|
|
"step": 9030
|
|
},
|
|
{
|
|
"entropy": 5.109751272201538,
|
|
"epoch": 9.10791729702471,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 6.064754903277078e-05,
|
|
"loss": 4.3359,
|
|
"mean_token_accuracy": 0.22939632833003998,
|
|
"num_tokens": 19955776.0,
|
|
"step": 9035
|
|
},
|
|
{
|
|
"entropy": 5.153336811065674,
|
|
"epoch": 9.112960161371658,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 6.052730268880027e-05,
|
|
"loss": 4.3479,
|
|
"mean_token_accuracy": 0.22787420451641083,
|
|
"num_tokens": 19966368.0,
|
|
"step": 9040
|
|
},
|
|
{
|
|
"entropy": 5.135177946090698,
|
|
"epoch": 9.118003025718608,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 6.040772293092049e-05,
|
|
"loss": 4.3685,
|
|
"mean_token_accuracy": 0.2371217802166939,
|
|
"num_tokens": 19978735.0,
|
|
"step": 9045
|
|
},
|
|
{
|
|
"entropy": 5.131320905685425,
|
|
"epoch": 9.123045890065557,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.0288810130787984e-05,
|
|
"loss": 4.3321,
|
|
"mean_token_accuracy": 0.22912597358226777,
|
|
"num_tokens": 19988754.0,
|
|
"step": 9050
|
|
},
|
|
{
|
|
"entropy": 5.165348100662231,
|
|
"epoch": 9.128088754412506,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 6.01705646579866e-05,
|
|
"loss": 4.3829,
|
|
"mean_token_accuracy": 0.2308268666267395,
|
|
"num_tokens": 19999492.0,
|
|
"step": 9055
|
|
},
|
|
{
|
|
"entropy": 5.208331441879272,
|
|
"epoch": 9.133131618759455,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.005298688002588e-05,
|
|
"loss": 4.4197,
|
|
"mean_token_accuracy": 0.22159984707832336,
|
|
"num_tokens": 20010059.0,
|
|
"step": 9060
|
|
},
|
|
{
|
|
"entropy": 5.127910947799682,
|
|
"epoch": 9.138174483106404,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 5.9936077162340316e-05,
|
|
"loss": 4.3466,
|
|
"mean_token_accuracy": 0.22905752509832383,
|
|
"num_tokens": 20020757.0,
|
|
"step": 9065
|
|
},
|
|
{
|
|
"entropy": 5.143463230133056,
|
|
"epoch": 9.143217347453353,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.981983586828807e-05,
|
|
"loss": 4.4267,
|
|
"mean_token_accuracy": 0.22525073885917662,
|
|
"num_tokens": 20033470.0,
|
|
"step": 9070
|
|
},
|
|
{
|
|
"entropy": 5.197110509872436,
|
|
"epoch": 9.148260211800302,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 5.970426335914969e-05,
|
|
"loss": 4.4703,
|
|
"mean_token_accuracy": 0.2209630072116852,
|
|
"num_tokens": 20044986.0,
|
|
"step": 9075
|
|
},
|
|
{
|
|
"entropy": 5.162522077560425,
|
|
"epoch": 9.153303076147251,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.9589359994127236e-05,
|
|
"loss": 4.3776,
|
|
"mean_token_accuracy": 0.22797417491674424,
|
|
"num_tokens": 20054694.0,
|
|
"step": 9080
|
|
},
|
|
{
|
|
"entropy": 5.12720799446106,
|
|
"epoch": 9.1583459404942,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.9475126130342996e-05,
|
|
"loss": 4.3916,
|
|
"mean_token_accuracy": 0.22684804052114488,
|
|
"num_tokens": 20066124.0,
|
|
"step": 9085
|
|
},
|
|
{
|
|
"entropy": 5.142040681838989,
|
|
"epoch": 9.16338880484115,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 5.936156212283845e-05,
|
|
"loss": 4.3777,
|
|
"mean_token_accuracy": 0.23573951572179794,
|
|
"num_tokens": 20077619.0,
|
|
"step": 9090
|
|
},
|
|
{
|
|
"entropy": 5.107533407211304,
|
|
"epoch": 9.168431669188099,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 5.924866832457317e-05,
|
|
"loss": 4.348,
|
|
"mean_token_accuracy": 0.23677872568368913,
|
|
"num_tokens": 20090350.0,
|
|
"step": 9095
|
|
},
|
|
{
|
|
"entropy": 5.181887626647949,
|
|
"epoch": 9.173474533535048,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 5.913644508642367e-05,
|
|
"loss": 4.3977,
|
|
"mean_token_accuracy": 0.23153521120548248,
|
|
"num_tokens": 20103120.0,
|
|
"step": 9100
|
|
},
|
|
{
|
|
"entropy": 5.196516513824463,
|
|
"epoch": 9.178517397881997,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.902489275718238e-05,
|
|
"loss": 4.3697,
|
|
"mean_token_accuracy": 0.23916791677474974,
|
|
"num_tokens": 20114488.0,
|
|
"step": 9105
|
|
},
|
|
{
|
|
"entropy": 5.160361194610596,
|
|
"epoch": 9.183560262228946,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 5.891401168355652e-05,
|
|
"loss": 4.3521,
|
|
"mean_token_accuracy": 0.24296543151140212,
|
|
"num_tokens": 20125006.0,
|
|
"step": 9110
|
|
},
|
|
{
|
|
"entropy": 5.090650415420532,
|
|
"epoch": 9.188603126575895,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 5.8803802210166994e-05,
|
|
"loss": 4.3771,
|
|
"mean_token_accuracy": 0.23016187697649002,
|
|
"num_tokens": 20135967.0,
|
|
"step": 9115
|
|
},
|
|
{
|
|
"entropy": 5.2296614170074465,
|
|
"epoch": 9.193645990922844,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 5.869426467954741e-05,
|
|
"loss": 4.3979,
|
|
"mean_token_accuracy": 0.22756254822015762,
|
|
"num_tokens": 20146670.0,
|
|
"step": 9120
|
|
},
|
|
{
|
|
"entropy": 5.095213079452515,
|
|
"epoch": 9.198688855269793,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 5.858539943214293e-05,
|
|
"loss": 4.3595,
|
|
"mean_token_accuracy": 0.23137693405151366,
|
|
"num_tokens": 20157818.0,
|
|
"step": 9125
|
|
},
|
|
{
|
|
"entropy": 5.148293113708496,
|
|
"epoch": 9.203731719616743,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 5.8477206806309236e-05,
|
|
"loss": 4.4304,
|
|
"mean_token_accuracy": 0.2218225598335266,
|
|
"num_tokens": 20169626.0,
|
|
"step": 9130
|
|
},
|
|
{
|
|
"entropy": 5.134908533096313,
|
|
"epoch": 9.208774583963692,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.8369687138311524e-05,
|
|
"loss": 4.3175,
|
|
"mean_token_accuracy": 0.2392900675535202,
|
|
"num_tokens": 20180015.0,
|
|
"step": 9135
|
|
},
|
|
{
|
|
"entropy": 5.119070196151734,
|
|
"epoch": 9.21381744831064,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 5.8262840762323435e-05,
|
|
"loss": 4.3493,
|
|
"mean_token_accuracy": 0.2296421781182289,
|
|
"num_tokens": 20192154.0,
|
|
"step": 9140
|
|
},
|
|
{
|
|
"entropy": 5.134674119949341,
|
|
"epoch": 9.21886031265759,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.815666801042593e-05,
|
|
"loss": 4.3588,
|
|
"mean_token_accuracy": 0.2327400490641594,
|
|
"num_tokens": 20202216.0,
|
|
"step": 9145
|
|
},
|
|
{
|
|
"entropy": 5.182086706161499,
|
|
"epoch": 9.223903177004539,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.8051169212606394e-05,
|
|
"loss": 4.3512,
|
|
"mean_token_accuracy": 0.234568090736866,
|
|
"num_tokens": 20212633.0,
|
|
"step": 9150
|
|
},
|
|
{
|
|
"entropy": 5.135794973373413,
|
|
"epoch": 9.228946041351488,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 5.794634469675749e-05,
|
|
"loss": 4.3735,
|
|
"mean_token_accuracy": 0.2234349638223648,
|
|
"num_tokens": 20223404.0,
|
|
"step": 9155
|
|
},
|
|
{
|
|
"entropy": 5.174224996566773,
|
|
"epoch": 9.233988905698437,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.7842194788676236e-05,
|
|
"loss": 4.4151,
|
|
"mean_token_accuracy": 0.2235453560948372,
|
|
"num_tokens": 20234756.0,
|
|
"step": 9160
|
|
},
|
|
{
|
|
"entropy": 5.160951900482178,
|
|
"epoch": 9.239031770045386,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 5.7738719812063e-05,
|
|
"loss": 4.351,
|
|
"mean_token_accuracy": 0.23868815153837203,
|
|
"num_tokens": 20245810.0,
|
|
"step": 9165
|
|
},
|
|
{
|
|
"entropy": 5.107413911819458,
|
|
"epoch": 9.244074634392335,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 5.763592008852036e-05,
|
|
"loss": 4.3596,
|
|
"mean_token_accuracy": 0.22982427030801772,
|
|
"num_tokens": 20256935.0,
|
|
"step": 9170
|
|
},
|
|
{
|
|
"entropy": 5.131717205047607,
|
|
"epoch": 9.249117498739285,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 5.753379593755221e-05,
|
|
"loss": 4.4129,
|
|
"mean_token_accuracy": 0.22574460804462432,
|
|
"num_tokens": 20268435.0,
|
|
"step": 9175
|
|
},
|
|
{
|
|
"entropy": 5.136482858657837,
|
|
"epoch": 9.254160363086234,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.743234767656277e-05,
|
|
"loss": 4.3561,
|
|
"mean_token_accuracy": 0.2312683254480362,
|
|
"num_tokens": 20278699.0,
|
|
"step": 9180
|
|
},
|
|
{
|
|
"entropy": 5.130120515823364,
|
|
"epoch": 9.259203227433183,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.733157562085565e-05,
|
|
"loss": 4.322,
|
|
"mean_token_accuracy": 0.23559814244508742,
|
|
"num_tokens": 20289457.0,
|
|
"step": 9185
|
|
},
|
|
{
|
|
"entropy": 5.144588279724121,
|
|
"epoch": 9.264246091780132,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 5.723148008363261e-05,
|
|
"loss": 4.3898,
|
|
"mean_token_accuracy": 0.227702796459198,
|
|
"num_tokens": 20300376.0,
|
|
"step": 9190
|
|
},
|
|
{
|
|
"entropy": 5.148876714706421,
|
|
"epoch": 9.269288956127081,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 5.713206137599295e-05,
|
|
"loss": 4.3752,
|
|
"mean_token_accuracy": 0.2335097998380661,
|
|
"num_tokens": 20310914.0,
|
|
"step": 9195
|
|
},
|
|
{
|
|
"entropy": 5.126593446731567,
|
|
"epoch": 9.274331820474028,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.703331980693233e-05,
|
|
"loss": 4.3602,
|
|
"mean_token_accuracy": 0.22521998435258866,
|
|
"num_tokens": 20321703.0,
|
|
"step": 9200
|
|
},
|
|
{
|
|
"entropy": 5.108891248703003,
|
|
"epoch": 9.279374684820977,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.69352556833418e-05,
|
|
"loss": 4.4056,
|
|
"mean_token_accuracy": 0.2255674868822098,
|
|
"num_tokens": 20331799.0,
|
|
"step": 9205
|
|
},
|
|
{
|
|
"entropy": 5.20147442817688,
|
|
"epoch": 9.284417549167927,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.6837869310006885e-05,
|
|
"loss": 4.4299,
|
|
"mean_token_accuracy": 0.2302386224269867,
|
|
"num_tokens": 20342471.0,
|
|
"step": 9210
|
|
},
|
|
{
|
|
"entropy": 5.158655261993408,
|
|
"epoch": 9.289460413514876,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.674116098960672e-05,
|
|
"loss": 4.3656,
|
|
"mean_token_accuracy": 0.2325798526406288,
|
|
"num_tokens": 20353204.0,
|
|
"step": 9215
|
|
},
|
|
{
|
|
"entropy": 5.156275606155395,
|
|
"epoch": 9.294503277861825,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.6645131022712974e-05,
|
|
"loss": 4.3856,
|
|
"mean_token_accuracy": 0.22989626675844194,
|
|
"num_tokens": 20363424.0,
|
|
"step": 9220
|
|
},
|
|
{
|
|
"entropy": 5.166053771972656,
|
|
"epoch": 9.299546142208774,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.6549779707789026e-05,
|
|
"loss": 4.3893,
|
|
"mean_token_accuracy": 0.22943311631679536,
|
|
"num_tokens": 20375804.0,
|
|
"step": 9225
|
|
},
|
|
{
|
|
"entropy": 5.1611387729644775,
|
|
"epoch": 9.304589006555723,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.645510734118892e-05,
|
|
"loss": 4.3725,
|
|
"mean_token_accuracy": 0.22811858355998993,
|
|
"num_tokens": 20385530.0,
|
|
"step": 9230
|
|
},
|
|
{
|
|
"entropy": 5.110063171386718,
|
|
"epoch": 9.309631870902672,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.636111421715662e-05,
|
|
"loss": 4.3003,
|
|
"mean_token_accuracy": 0.2406669095158577,
|
|
"num_tokens": 20395981.0,
|
|
"step": 9235
|
|
},
|
|
{
|
|
"entropy": 5.165987634658814,
|
|
"epoch": 9.314674735249621,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.6267800627824856e-05,
|
|
"loss": 4.3822,
|
|
"mean_token_accuracy": 0.22826223075389862,
|
|
"num_tokens": 20406685.0,
|
|
"step": 9240
|
|
},
|
|
{
|
|
"entropy": 5.144363594055176,
|
|
"epoch": 9.31971759959657,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.617516686321446e-05,
|
|
"loss": 4.3682,
|
|
"mean_token_accuracy": 0.2303505763411522,
|
|
"num_tokens": 20416554.0,
|
|
"step": 9245
|
|
},
|
|
{
|
|
"entropy": 5.14738826751709,
|
|
"epoch": 9.32476046394352,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 5.608321321123328e-05,
|
|
"loss": 4.3748,
|
|
"mean_token_accuracy": 0.2380754381418228,
|
|
"num_tokens": 20427755.0,
|
|
"step": 9250
|
|
},
|
|
{
|
|
"entropy": 5.138025760650635,
|
|
"epoch": 9.329803328290469,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.59919399576754e-05,
|
|
"loss": 4.3616,
|
|
"mean_token_accuracy": 0.23061030805110933,
|
|
"num_tokens": 20438975.0,
|
|
"step": 9255
|
|
},
|
|
{
|
|
"entropy": 5.13541693687439,
|
|
"epoch": 9.334846192637418,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.5901347386220205e-05,
|
|
"loss": 4.359,
|
|
"mean_token_accuracy": 0.23097197711467743,
|
|
"num_tokens": 20449579.0,
|
|
"step": 9260
|
|
},
|
|
{
|
|
"entropy": 5.056546831130982,
|
|
"epoch": 9.339889056984367,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 5.581143577843152e-05,
|
|
"loss": 4.2809,
|
|
"mean_token_accuracy": 0.23605398386716842,
|
|
"num_tokens": 20460642.0,
|
|
"step": 9265
|
|
},
|
|
{
|
|
"entropy": 5.137201881408691,
|
|
"epoch": 9.344931921331316,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 5.5722205413756655e-05,
|
|
"loss": 4.3451,
|
|
"mean_token_accuracy": 0.22792994379997253,
|
|
"num_tokens": 20471059.0,
|
|
"step": 9270
|
|
},
|
|
{
|
|
"entropy": 5.084278917312622,
|
|
"epoch": 9.349974785678265,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.563365656952563e-05,
|
|
"loss": 4.3127,
|
|
"mean_token_accuracy": 0.24630985110998155,
|
|
"num_tokens": 20481075.0,
|
|
"step": 9275
|
|
},
|
|
{
|
|
"entropy": 5.187985801696778,
|
|
"epoch": 9.355017650025214,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.5545789520950324e-05,
|
|
"loss": 4.3908,
|
|
"mean_token_accuracy": 0.22502024620771408,
|
|
"num_tokens": 20491094.0,
|
|
"step": 9280
|
|
},
|
|
{
|
|
"entropy": 5.056336545944214,
|
|
"epoch": 9.360060514372163,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.545860454112355e-05,
|
|
"loss": 4.2653,
|
|
"mean_token_accuracy": 0.24398380219936372,
|
|
"num_tokens": 20501960.0,
|
|
"step": 9285
|
|
},
|
|
{
|
|
"entropy": 5.12333345413208,
|
|
"epoch": 9.365103378719112,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.537210190101825e-05,
|
|
"loss": 4.3352,
|
|
"mean_token_accuracy": 0.238752843439579,
|
|
"num_tokens": 20512416.0,
|
|
"step": 9290
|
|
},
|
|
{
|
|
"entropy": 5.068557024002075,
|
|
"epoch": 9.370146243066062,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 5.5286281869486585e-05,
|
|
"loss": 4.2984,
|
|
"mean_token_accuracy": 0.23333841562271118,
|
|
"num_tokens": 20523384.0,
|
|
"step": 9295
|
|
},
|
|
{
|
|
"entropy": 5.130578947067261,
|
|
"epoch": 9.37518910741301,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 5.5201144713259234e-05,
|
|
"loss": 4.408,
|
|
"mean_token_accuracy": 0.23150297701358796,
|
|
"num_tokens": 20535718.0,
|
|
"step": 9300
|
|
},
|
|
{
|
|
"entropy": 5.129572582244873,
|
|
"epoch": 9.38023197175996,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 5.511669069694441e-05,
|
|
"loss": 4.3234,
|
|
"mean_token_accuracy": 0.2349911704659462,
|
|
"num_tokens": 20547155.0,
|
|
"step": 9305
|
|
},
|
|
{
|
|
"entropy": 5.13906683921814,
|
|
"epoch": 9.385274836106909,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 5.503292008302719e-05,
|
|
"loss": 4.3928,
|
|
"mean_token_accuracy": 0.2263153687119484,
|
|
"num_tokens": 20558889.0,
|
|
"step": 9310
|
|
},
|
|
{
|
|
"entropy": 5.175556182861328,
|
|
"epoch": 9.390317700453858,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 5.494983313186849e-05,
|
|
"loss": 4.3954,
|
|
"mean_token_accuracy": 0.24123041927814484,
|
|
"num_tokens": 20569604.0,
|
|
"step": 9315
|
|
},
|
|
{
|
|
"entropy": 5.17207932472229,
|
|
"epoch": 9.395360564800807,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.4867430101704565e-05,
|
|
"loss": 4.3947,
|
|
"mean_token_accuracy": 0.23149098008871077,
|
|
"num_tokens": 20580473.0,
|
|
"step": 9320
|
|
},
|
|
{
|
|
"entropy": 5.126982402801514,
|
|
"epoch": 9.400403429147756,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.4785711248645854e-05,
|
|
"loss": 4.4153,
|
|
"mean_token_accuracy": 0.2257186710834503,
|
|
"num_tokens": 20591023.0,
|
|
"step": 9325
|
|
},
|
|
{
|
|
"entropy": 5.214831256866455,
|
|
"epoch": 9.405446293494705,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 5.470467682667653e-05,
|
|
"loss": 4.4587,
|
|
"mean_token_accuracy": 0.22180478572845458,
|
|
"num_tokens": 20603062.0,
|
|
"step": 9330
|
|
},
|
|
{
|
|
"entropy": 5.173099565505981,
|
|
"epoch": 9.410489157841655,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 5.462432708765333e-05,
|
|
"loss": 4.3955,
|
|
"mean_token_accuracy": 0.22516263872385026,
|
|
"num_tokens": 20613849.0,
|
|
"step": 9335
|
|
},
|
|
{
|
|
"entropy": 5.2046473026275635,
|
|
"epoch": 9.415532022188604,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 5.4544662281305214e-05,
|
|
"loss": 4.3727,
|
|
"mean_token_accuracy": 0.23650218099355697,
|
|
"num_tokens": 20625039.0,
|
|
"step": 9340
|
|
},
|
|
{
|
|
"entropy": 5.168882656097412,
|
|
"epoch": 9.420574886535553,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 5.446568265523215e-05,
|
|
"loss": 4.3523,
|
|
"mean_token_accuracy": 0.22802196443080902,
|
|
"num_tokens": 20636233.0,
|
|
"step": 9345
|
|
},
|
|
{
|
|
"entropy": 5.1501518249511715,
|
|
"epoch": 9.425617750882502,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 5.4387388454904656e-05,
|
|
"loss": 4.3629,
|
|
"mean_token_accuracy": 0.22581951469182968,
|
|
"num_tokens": 20647100.0,
|
|
"step": 9350
|
|
},
|
|
{
|
|
"entropy": 5.16693377494812,
|
|
"epoch": 9.430660615229451,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 5.4309779923662956e-05,
|
|
"loss": 4.4515,
|
|
"mean_token_accuracy": 0.21709060668945312,
|
|
"num_tokens": 20659505.0,
|
|
"step": 9355
|
|
},
|
|
{
|
|
"entropy": 5.120960521697998,
|
|
"epoch": 9.4357034795764,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.423285730271612e-05,
|
|
"loss": 4.3427,
|
|
"mean_token_accuracy": 0.22867295444011687,
|
|
"num_tokens": 20669081.0,
|
|
"step": 9360
|
|
},
|
|
{
|
|
"entropy": 5.1185564517974855,
|
|
"epoch": 9.44074634392335,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 5.415662083114142e-05,
|
|
"loss": 4.3611,
|
|
"mean_token_accuracy": 0.2308626726269722,
|
|
"num_tokens": 20679421.0,
|
|
"step": 9365
|
|
},
|
|
{
|
|
"entropy": 5.117571830749512,
|
|
"epoch": 9.445789208270298,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.4081070745883584e-05,
|
|
"loss": 4.3658,
|
|
"mean_token_accuracy": 0.22271402031183243,
|
|
"num_tokens": 20689884.0,
|
|
"step": 9370
|
|
},
|
|
{
|
|
"entropy": 5.146344757080078,
|
|
"epoch": 9.450832072617247,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 5.4006207281754035e-05,
|
|
"loss": 4.3871,
|
|
"mean_token_accuracy": 0.22268199920654297,
|
|
"num_tokens": 20701161.0,
|
|
"step": 9375
|
|
},
|
|
{
|
|
"entropy": 5.147802686691284,
|
|
"epoch": 9.455874936964197,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 5.393203067143011e-05,
|
|
"loss": 4.4305,
|
|
"mean_token_accuracy": 0.22093244940042495,
|
|
"num_tokens": 20712854.0,
|
|
"step": 9380
|
|
},
|
|
{
|
|
"entropy": 5.123865270614624,
|
|
"epoch": 9.460917801311144,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.385854114545445e-05,
|
|
"loss": 4.3153,
|
|
"mean_token_accuracy": 0.2424059733748436,
|
|
"num_tokens": 20723214.0,
|
|
"step": 9385
|
|
},
|
|
{
|
|
"entropy": 5.168157434463501,
|
|
"epoch": 9.465960665658093,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.3785738932234175e-05,
|
|
"loss": 4.351,
|
|
"mean_token_accuracy": 0.23127028048038484,
|
|
"num_tokens": 20734912.0,
|
|
"step": 9390
|
|
},
|
|
{
|
|
"entropy": 5.14145975112915,
|
|
"epoch": 9.471003530005042,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.3713624258040286e-05,
|
|
"loss": 4.309,
|
|
"mean_token_accuracy": 0.23687906712293624,
|
|
"num_tokens": 20745377.0,
|
|
"step": 9395
|
|
},
|
|
{
|
|
"entropy": 5.064557790756226,
|
|
"epoch": 9.476046394351991,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 5.364219734700682e-05,
|
|
"loss": 4.2482,
|
|
"mean_token_accuracy": 0.2424289330840111,
|
|
"num_tokens": 20755778.0,
|
|
"step": 9400
|
|
},
|
|
{
|
|
"entropy": 5.124008560180664,
|
|
"epoch": 9.48108925869894,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 5.357145842113027e-05,
|
|
"loss": 4.4364,
|
|
"mean_token_accuracy": 0.22049185484647751,
|
|
"num_tokens": 20768625.0,
|
|
"step": 9405
|
|
},
|
|
{
|
|
"entropy": 5.141441011428833,
|
|
"epoch": 9.48613212304589,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 5.350140770026885e-05,
|
|
"loss": 4.3565,
|
|
"mean_token_accuracy": 0.23239268958568574,
|
|
"num_tokens": 20779562.0,
|
|
"step": 9410
|
|
},
|
|
{
|
|
"entropy": 5.1605321884155275,
|
|
"epoch": 9.491174987392839,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.34320454021418e-05,
|
|
"loss": 4.317,
|
|
"mean_token_accuracy": 0.23492688089609146,
|
|
"num_tokens": 20790181.0,
|
|
"step": 9415
|
|
},
|
|
{
|
|
"entropy": 5.113265943527222,
|
|
"epoch": 9.496217851739788,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 5.336337174232878e-05,
|
|
"loss": 4.411,
|
|
"mean_token_accuracy": 0.22292036861181258,
|
|
"num_tokens": 20802401.0,
|
|
"step": 9420
|
|
},
|
|
{
|
|
"entropy": 5.124378442764282,
|
|
"epoch": 9.501260716086737,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.329538693426908e-05,
|
|
"loss": 4.3663,
|
|
"mean_token_accuracy": 0.22928679138422012,
|
|
"num_tokens": 20814980.0,
|
|
"step": 9425
|
|
},
|
|
{
|
|
"entropy": 5.137832689285278,
|
|
"epoch": 9.506303580433686,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 5.322809118926106e-05,
|
|
"loss": 4.3632,
|
|
"mean_token_accuracy": 0.2220934510231018,
|
|
"num_tokens": 20825996.0,
|
|
"step": 9430
|
|
},
|
|
{
|
|
"entropy": 5.172269630432129,
|
|
"epoch": 9.511346444780635,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.316148471646147e-05,
|
|
"loss": 4.4428,
|
|
"mean_token_accuracy": 0.2274964839220047,
|
|
"num_tokens": 20837176.0,
|
|
"step": 9435
|
|
},
|
|
{
|
|
"entropy": 5.1667149543762205,
|
|
"epoch": 9.516389309127584,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.30955677228847e-05,
|
|
"loss": 4.4345,
|
|
"mean_token_accuracy": 0.22802691608667375,
|
|
"num_tokens": 20847698.0,
|
|
"step": 9440
|
|
},
|
|
{
|
|
"entropy": 5.1788482666015625,
|
|
"epoch": 9.521432173474533,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.303034041340237e-05,
|
|
"loss": 4.3901,
|
|
"mean_token_accuracy": 0.22766114622354508,
|
|
"num_tokens": 20858269.0,
|
|
"step": 9445
|
|
},
|
|
{
|
|
"entropy": 5.130408477783203,
|
|
"epoch": 9.526475037821482,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 5.296580299074239e-05,
|
|
"loss": 4.4117,
|
|
"mean_token_accuracy": 0.21768154203891754,
|
|
"num_tokens": 20869553.0,
|
|
"step": 9450
|
|
},
|
|
{
|
|
"entropy": 5.082242774963379,
|
|
"epoch": 9.531517902168432,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.2901955655488616e-05,
|
|
"loss": 4.2854,
|
|
"mean_token_accuracy": 0.23998129665851592,
|
|
"num_tokens": 20880220.0,
|
|
"step": 9455
|
|
},
|
|
{
|
|
"entropy": 5.150358104705811,
|
|
"epoch": 9.53656076651538,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 5.283879860607998e-05,
|
|
"loss": 4.3478,
|
|
"mean_token_accuracy": 0.22902153581380844,
|
|
"num_tokens": 20890603.0,
|
|
"step": 9460
|
|
},
|
|
{
|
|
"entropy": 5.169724941253662,
|
|
"epoch": 9.54160363086233,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 5.277633203881011e-05,
|
|
"loss": 4.4441,
|
|
"mean_token_accuracy": 0.22400763332843782,
|
|
"num_tokens": 20902069.0,
|
|
"step": 9465
|
|
},
|
|
{
|
|
"entropy": 5.153227138519287,
|
|
"epoch": 9.546646495209279,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.271455614782649e-05,
|
|
"loss": 4.3936,
|
|
"mean_token_accuracy": 0.2287513419985771,
|
|
"num_tokens": 20912936.0,
|
|
"step": 9470
|
|
},
|
|
{
|
|
"entropy": 5.133691072463989,
|
|
"epoch": 9.551689359556228,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 5.265347112513003e-05,
|
|
"loss": 4.3389,
|
|
"mean_token_accuracy": 0.2301792547106743,
|
|
"num_tokens": 20924877.0,
|
|
"step": 9475
|
|
},
|
|
{
|
|
"entropy": 5.140048170089722,
|
|
"epoch": 9.556732223903177,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.2593077160574414e-05,
|
|
"loss": 4.4092,
|
|
"mean_token_accuracy": 0.22877945899963378,
|
|
"num_tokens": 20934759.0,
|
|
"step": 9480
|
|
},
|
|
{
|
|
"entropy": 5.104368495941162,
|
|
"epoch": 9.561775088250126,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 5.2533374441865447e-05,
|
|
"loss": 4.3243,
|
|
"mean_token_accuracy": 0.23089100122451783,
|
|
"num_tokens": 20945894.0,
|
|
"step": 9485
|
|
},
|
|
{
|
|
"entropy": 5.140207815170288,
|
|
"epoch": 9.566817952597075,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.247436315456057e-05,
|
|
"loss": 4.3339,
|
|
"mean_token_accuracy": 0.23620492666959764,
|
|
"num_tokens": 20956720.0,
|
|
"step": 9490
|
|
},
|
|
{
|
|
"entropy": 5.128748655319214,
|
|
"epoch": 9.571860816944024,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.2416043482068214e-05,
|
|
"loss": 4.329,
|
|
"mean_token_accuracy": 0.23562139570713042,
|
|
"num_tokens": 20967097.0,
|
|
"step": 9495
|
|
},
|
|
{
|
|
"entropy": 5.206929683685303,
|
|
"epoch": 9.576903681290974,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.2358415605647264e-05,
|
|
"loss": 4.4263,
|
|
"mean_token_accuracy": 0.22582029402256013,
|
|
"num_tokens": 20977727.0,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"epoch": 9.576903681290974,
|
|
"eval_entropy": 5.090632214194339,
|
|
"eval_loss": 5.697966575622559,
|
|
"eval_mean_token_accuracy": 0.15582988320188668,
|
|
"eval_num_tokens": 20977727.0,
|
|
"eval_runtime": 2.8484,
|
|
"eval_samples_per_second": 1101.66,
|
|
"eval_steps_per_second": 137.971,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"entropy": 5.135790061950684,
|
|
"epoch": 9.581946545637923,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 5.2301479704406514e-05,
|
|
"loss": 4.3782,
|
|
"mean_token_accuracy": 0.23183967918157578,
|
|
"num_tokens": 20987604.0,
|
|
"step": 9505
|
|
},
|
|
{
|
|
"entropy": 5.146976327896118,
|
|
"epoch": 9.586989409984872,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.224523595530402e-05,
|
|
"loss": 4.445,
|
|
"mean_token_accuracy": 0.21637622117996216,
|
|
"num_tokens": 20999351.0,
|
|
"step": 9510
|
|
},
|
|
{
|
|
"entropy": 5.168024635314941,
|
|
"epoch": 9.592032274331821,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.2189684533146676e-05,
|
|
"loss": 4.4491,
|
|
"mean_token_accuracy": 0.22097799479961394,
|
|
"num_tokens": 21010645.0,
|
|
"step": 9515
|
|
},
|
|
{
|
|
"entropy": 5.077105140686035,
|
|
"epoch": 9.59707513867877,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 5.213482561058962e-05,
|
|
"loss": 4.341,
|
|
"mean_token_accuracy": 0.2363715261220932,
|
|
"num_tokens": 21021300.0,
|
|
"step": 9520
|
|
},
|
|
{
|
|
"entropy": 5.1505101203918455,
|
|
"epoch": 9.60211800302572,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 5.208065935813556e-05,
|
|
"loss": 4.3662,
|
|
"mean_token_accuracy": 0.23421839028596877,
|
|
"num_tokens": 21032214.0,
|
|
"step": 9525
|
|
},
|
|
{
|
|
"entropy": 5.12214617729187,
|
|
"epoch": 9.607160867372668,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 5.2027185944134535e-05,
|
|
"loss": 4.3388,
|
|
"mean_token_accuracy": 0.23574193120002745,
|
|
"num_tokens": 21042411.0,
|
|
"step": 9530
|
|
},
|
|
{
|
|
"entropy": 5.154958724975586,
|
|
"epoch": 9.612203731719617,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 5.197440553478312e-05,
|
|
"loss": 4.3693,
|
|
"mean_token_accuracy": 0.23050934374332427,
|
|
"num_tokens": 21053692.0,
|
|
"step": 9535
|
|
},
|
|
{
|
|
"entropy": 5.146642684936523,
|
|
"epoch": 9.617246596066567,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.192231829412404e-05,
|
|
"loss": 4.3965,
|
|
"mean_token_accuracy": 0.2343198761343956,
|
|
"num_tokens": 21064817.0,
|
|
"step": 9540
|
|
},
|
|
{
|
|
"entropy": 5.10263500213623,
|
|
"epoch": 9.622289460413516,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 5.187092438404565e-05,
|
|
"loss": 4.3222,
|
|
"mean_token_accuracy": 0.2376010686159134,
|
|
"num_tokens": 21077592.0,
|
|
"step": 9545
|
|
},
|
|
{
|
|
"entropy": 5.1478558540344235,
|
|
"epoch": 9.627332324760465,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 5.18202239642814e-05,
|
|
"loss": 4.3883,
|
|
"mean_token_accuracy": 0.23269044011831283,
|
|
"num_tokens": 21089386.0,
|
|
"step": 9550
|
|
},
|
|
{
|
|
"entropy": 5.134736824035644,
|
|
"epoch": 9.632375189107414,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 5.177021719240939e-05,
|
|
"loss": 4.3724,
|
|
"mean_token_accuracy": 0.22660518288612366,
|
|
"num_tokens": 21101090.0,
|
|
"step": 9555
|
|
},
|
|
{
|
|
"entropy": 5.185186958312988,
|
|
"epoch": 9.637418053454361,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 5.172090422385178e-05,
|
|
"loss": 4.3226,
|
|
"mean_token_accuracy": 0.23731129616498947,
|
|
"num_tokens": 21111958.0,
|
|
"step": 9560
|
|
},
|
|
{
|
|
"entropy": 5.1961729526519775,
|
|
"epoch": 9.64246091780131,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 5.167228521187443e-05,
|
|
"loss": 4.3979,
|
|
"mean_token_accuracy": 0.2301601454615593,
|
|
"num_tokens": 21123086.0,
|
|
"step": 9565
|
|
},
|
|
{
|
|
"entropy": 5.188040637969971,
|
|
"epoch": 9.64750378214826,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 5.162436030758637e-05,
|
|
"loss": 4.3825,
|
|
"mean_token_accuracy": 0.23821073174476623,
|
|
"num_tokens": 21134374.0,
|
|
"step": 9570
|
|
},
|
|
{
|
|
"entropy": 5.1548659324646,
|
|
"epoch": 9.652546646495209,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 5.1577129659939304e-05,
|
|
"loss": 4.4193,
|
|
"mean_token_accuracy": 0.23094950765371322,
|
|
"num_tokens": 21145905.0,
|
|
"step": 9575
|
|
},
|
|
{
|
|
"entropy": 5.130598497390747,
|
|
"epoch": 9.657589510842158,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.1530593415727143e-05,
|
|
"loss": 4.3394,
|
|
"mean_token_accuracy": 0.23124466240406036,
|
|
"num_tokens": 21156131.0,
|
|
"step": 9580
|
|
},
|
|
{
|
|
"entropy": 5.147165060043335,
|
|
"epoch": 9.662632375189107,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 5.148475171958564e-05,
|
|
"loss": 4.4158,
|
|
"mean_token_accuracy": 0.23793899416923522,
|
|
"num_tokens": 21167644.0,
|
|
"step": 9585
|
|
},
|
|
{
|
|
"entropy": 5.158491945266723,
|
|
"epoch": 9.667675239536056,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.143960471399182e-05,
|
|
"loss": 4.4029,
|
|
"mean_token_accuracy": 0.22647368013858796,
|
|
"num_tokens": 21177175.0,
|
|
"step": 9590
|
|
},
|
|
{
|
|
"entropy": 5.176330614089966,
|
|
"epoch": 9.672718103883005,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 5.139515253926365e-05,
|
|
"loss": 4.3595,
|
|
"mean_token_accuracy": 0.22564810365438462,
|
|
"num_tokens": 21187367.0,
|
|
"step": 9595
|
|
},
|
|
{
|
|
"entropy": 5.143256950378418,
|
|
"epoch": 9.677760968229954,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 5.135139533355947e-05,
|
|
"loss": 4.4326,
|
|
"mean_token_accuracy": 0.22299969494342803,
|
|
"num_tokens": 21198193.0,
|
|
"step": 9600
|
|
},
|
|
{
|
|
"entropy": 5.145635986328125,
|
|
"epoch": 9.682803832576903,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.1308333232877724e-05,
|
|
"loss": 4.3709,
|
|
"mean_token_accuracy": 0.23549260795116425,
|
|
"num_tokens": 21210345.0,
|
|
"step": 9605
|
|
},
|
|
{
|
|
"entropy": 5.191702890396118,
|
|
"epoch": 9.687846696923852,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.1265966371056376e-05,
|
|
"loss": 4.449,
|
|
"mean_token_accuracy": 0.22445696741342544,
|
|
"num_tokens": 21222670.0,
|
|
"step": 9610
|
|
},
|
|
{
|
|
"entropy": 5.178430557250977,
|
|
"epoch": 9.692889561270801,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 5.1224294879772655e-05,
|
|
"loss": 4.3735,
|
|
"mean_token_accuracy": 0.2291101559996605,
|
|
"num_tokens": 21233199.0,
|
|
"step": 9615
|
|
},
|
|
{
|
|
"entropy": 5.1656121730804445,
|
|
"epoch": 9.69793242561775,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 5.118331888854251e-05,
|
|
"loss": 4.3502,
|
|
"mean_token_accuracy": 0.23833866268396378,
|
|
"num_tokens": 21243600.0,
|
|
"step": 9620
|
|
},
|
|
{
|
|
"entropy": 5.11466908454895,
|
|
"epoch": 9.7029752899647,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.114303852472023e-05,
|
|
"loss": 4.3477,
|
|
"mean_token_accuracy": 0.23547186106443405,
|
|
"num_tokens": 21253329.0,
|
|
"step": 9625
|
|
},
|
|
{
|
|
"entropy": 5.0902492046356205,
|
|
"epoch": 9.708018154311649,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 5.1103453913498126e-05,
|
|
"loss": 4.3474,
|
|
"mean_token_accuracy": 0.23052302300930022,
|
|
"num_tokens": 21263357.0,
|
|
"step": 9630
|
|
},
|
|
{
|
|
"entropy": 5.140036582946777,
|
|
"epoch": 9.713061018658598,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.106456517790611e-05,
|
|
"loss": 4.3462,
|
|
"mean_token_accuracy": 0.24162171334028243,
|
|
"num_tokens": 21274831.0,
|
|
"step": 9635
|
|
},
|
|
{
|
|
"entropy": 5.254909658432007,
|
|
"epoch": 9.718103883005547,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.1026372438811246e-05,
|
|
"loss": 4.5161,
|
|
"mean_token_accuracy": 0.21778088808059692,
|
|
"num_tokens": 21286687.0,
|
|
"step": 9640
|
|
},
|
|
{
|
|
"entropy": 5.125511932373047,
|
|
"epoch": 9.723146747352496,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.098887581491747e-05,
|
|
"loss": 4.3154,
|
|
"mean_token_accuracy": 0.24008737355470658,
|
|
"num_tokens": 21297063.0,
|
|
"step": 9645
|
|
},
|
|
{
|
|
"entropy": 5.129090261459351,
|
|
"epoch": 9.728189611699445,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 5.095207542276513e-05,
|
|
"loss": 4.3665,
|
|
"mean_token_accuracy": 0.2283402755856514,
|
|
"num_tokens": 21308446.0,
|
|
"step": 9650
|
|
},
|
|
{
|
|
"entropy": 5.15674843788147,
|
|
"epoch": 9.733232476046394,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 5.091597137673073e-05,
|
|
"loss": 4.4143,
|
|
"mean_token_accuracy": 0.22495699375867845,
|
|
"num_tokens": 21320572.0,
|
|
"step": 9655
|
|
},
|
|
{
|
|
"entropy": 5.152351951599121,
|
|
"epoch": 9.738275340393344,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 5.088056378902641e-05,
|
|
"loss": 4.4412,
|
|
"mean_token_accuracy": 0.22398223131895065,
|
|
"num_tokens": 21332678.0,
|
|
"step": 9660
|
|
},
|
|
{
|
|
"entropy": 5.140425491333008,
|
|
"epoch": 9.743318204740293,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.0845852769699824e-05,
|
|
"loss": 4.3745,
|
|
"mean_token_accuracy": 0.23215449303388597,
|
|
"num_tokens": 21343391.0,
|
|
"step": 9665
|
|
},
|
|
{
|
|
"entropy": 5.213333702087402,
|
|
"epoch": 9.748361069087242,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 5.081183842663361e-05,
|
|
"loss": 4.4248,
|
|
"mean_token_accuracy": 0.22803296148777008,
|
|
"num_tokens": 21355109.0,
|
|
"step": 9670
|
|
},
|
|
{
|
|
"entropy": 5.1526954650878904,
|
|
"epoch": 9.75340393343419,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 5.0778520865545154e-05,
|
|
"loss": 4.3999,
|
|
"mean_token_accuracy": 0.22909243553876876,
|
|
"num_tokens": 21367599.0,
|
|
"step": 9675
|
|
},
|
|
{
|
|
"entropy": 5.1555821895599365,
|
|
"epoch": 9.75844679778114,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 5.074590018998614e-05,
|
|
"loss": 4.401,
|
|
"mean_token_accuracy": 0.2328573152422905,
|
|
"num_tokens": 21378912.0,
|
|
"step": 9680
|
|
},
|
|
{
|
|
"entropy": 5.163139390945434,
|
|
"epoch": 9.763489662128089,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 5.071397650134244e-05,
|
|
"loss": 4.3269,
|
|
"mean_token_accuracy": 0.23135206699371338,
|
|
"num_tokens": 21390280.0,
|
|
"step": 9685
|
|
},
|
|
{
|
|
"entropy": 5.167592239379883,
|
|
"epoch": 9.768532526475038,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.0682749898833575e-05,
|
|
"loss": 4.4014,
|
|
"mean_token_accuracy": 0.22541940063238144,
|
|
"num_tokens": 21401964.0,
|
|
"step": 9690
|
|
},
|
|
{
|
|
"entropy": 5.193075275421142,
|
|
"epoch": 9.773575390821987,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 5.065222047951257e-05,
|
|
"loss": 4.411,
|
|
"mean_token_accuracy": 0.22676493376493453,
|
|
"num_tokens": 21412622.0,
|
|
"step": 9695
|
|
},
|
|
{
|
|
"entropy": 5.131914663314819,
|
|
"epoch": 9.778618255168936,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.0622388338265574e-05,
|
|
"loss": 4.3434,
|
|
"mean_token_accuracy": 0.2325545534491539,
|
|
"num_tokens": 21423984.0,
|
|
"step": 9700
|
|
},
|
|
{
|
|
"entropy": 5.172040605545044,
|
|
"epoch": 9.783661119515886,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.059325356781154e-05,
|
|
"loss": 4.4177,
|
|
"mean_token_accuracy": 0.22562895566225052,
|
|
"num_tokens": 21435393.0,
|
|
"step": 9705
|
|
},
|
|
{
|
|
"entropy": 5.146768188476562,
|
|
"epoch": 9.788703983862835,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 5.0564816258702034e-05,
|
|
"loss": 4.4321,
|
|
"mean_token_accuracy": 0.22684621810913086,
|
|
"num_tokens": 21445969.0,
|
|
"step": 9710
|
|
},
|
|
{
|
|
"entropy": 5.128154277801514,
|
|
"epoch": 9.793746848209784,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.053707649932085e-05,
|
|
"loss": 4.4086,
|
|
"mean_token_accuracy": 0.22969648838043213,
|
|
"num_tokens": 21457847.0,
|
|
"step": 9715
|
|
},
|
|
{
|
|
"entropy": 5.159723949432373,
|
|
"epoch": 9.798789712556733,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 5.0510034375883815e-05,
|
|
"loss": 4.419,
|
|
"mean_token_accuracy": 0.2307723268866539,
|
|
"num_tokens": 21469857.0,
|
|
"step": 9720
|
|
},
|
|
{
|
|
"entropy": 5.215474843978882,
|
|
"epoch": 9.803832576903682,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.0483689972438454e-05,
|
|
"loss": 4.412,
|
|
"mean_token_accuracy": 0.22089962065219879,
|
|
"num_tokens": 21480698.0,
|
|
"step": 9725
|
|
},
|
|
{
|
|
"entropy": 5.155325984954834,
|
|
"epoch": 9.808875441250631,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 5.045804337086378e-05,
|
|
"loss": 4.3473,
|
|
"mean_token_accuracy": 0.23258139342069625,
|
|
"num_tokens": 21491583.0,
|
|
"step": 9730
|
|
},
|
|
{
|
|
"entropy": 5.176821899414063,
|
|
"epoch": 9.813918305597578,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.043309465087006e-05,
|
|
"loss": 4.3797,
|
|
"mean_token_accuracy": 0.2303215429186821,
|
|
"num_tokens": 21502164.0,
|
|
"step": 9735
|
|
},
|
|
{
|
|
"entropy": 5.179122638702393,
|
|
"epoch": 9.81896116994453,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.040884388999846e-05,
|
|
"loss": 4.3811,
|
|
"mean_token_accuracy": 0.23362878412008287,
|
|
"num_tokens": 21513239.0,
|
|
"step": 9740
|
|
},
|
|
{
|
|
"entropy": 5.173216009140015,
|
|
"epoch": 9.824004034291477,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 5.0385291163620896e-05,
|
|
"loss": 4.4071,
|
|
"mean_token_accuracy": 0.2262898474931717,
|
|
"num_tokens": 21524928.0,
|
|
"step": 9745
|
|
},
|
|
{
|
|
"entropy": 5.157050561904907,
|
|
"epoch": 9.829046898638426,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 5.03624365449398e-05,
|
|
"loss": 4.3992,
|
|
"mean_token_accuracy": 0.230339352786541,
|
|
"num_tokens": 21536912.0,
|
|
"step": 9750
|
|
},
|
|
{
|
|
"entropy": 5.164587593078613,
|
|
"epoch": 9.834089762985375,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 5.034028010498784e-05,
|
|
"loss": 4.3798,
|
|
"mean_token_accuracy": 0.22676306515932082,
|
|
"num_tokens": 21548271.0,
|
|
"step": 9755
|
|
},
|
|
{
|
|
"entropy": 5.125851106643677,
|
|
"epoch": 9.839132627332324,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 5.0318821912627776e-05,
|
|
"loss": 4.2975,
|
|
"mean_token_accuracy": 0.23196963369846343,
|
|
"num_tokens": 21558863.0,
|
|
"step": 9760
|
|
},
|
|
{
|
|
"entropy": 5.12449688911438,
|
|
"epoch": 9.844175491679273,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.029806203455213e-05,
|
|
"loss": 4.3418,
|
|
"mean_token_accuracy": 0.22532624155282974,
|
|
"num_tokens": 21569957.0,
|
|
"step": 9765
|
|
},
|
|
{
|
|
"entropy": 5.118330383300782,
|
|
"epoch": 9.849218356026222,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 5.027800053528309e-05,
|
|
"loss": 4.33,
|
|
"mean_token_accuracy": 0.2294528990983963,
|
|
"num_tokens": 21580238.0,
|
|
"step": 9770
|
|
},
|
|
{
|
|
"entropy": 5.1732433319091795,
|
|
"epoch": 9.854261220373171,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.025863747717227e-05,
|
|
"loss": 4.3674,
|
|
"mean_token_accuracy": 0.22685556560754777,
|
|
"num_tokens": 21591154.0,
|
|
"step": 9775
|
|
},
|
|
{
|
|
"entropy": 5.168726301193237,
|
|
"epoch": 9.85930408472012,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 5.0239972920400485e-05,
|
|
"loss": 4.3808,
|
|
"mean_token_accuracy": 0.2290065824985504,
|
|
"num_tokens": 21603095.0,
|
|
"step": 9780
|
|
},
|
|
{
|
|
"entropy": 5.105444145202637,
|
|
"epoch": 9.86434694906707,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.022200692297763e-05,
|
|
"loss": 4.3653,
|
|
"mean_token_accuracy": 0.23141383081674577,
|
|
"num_tokens": 21614401.0,
|
|
"step": 9785
|
|
},
|
|
{
|
|
"entropy": 5.173939895629883,
|
|
"epoch": 9.869389813414019,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 5.0204739540742436e-05,
|
|
"loss": 4.3683,
|
|
"mean_token_accuracy": 0.23142160922288896,
|
|
"num_tokens": 21625037.0,
|
|
"step": 9790
|
|
},
|
|
{
|
|
"entropy": 5.181688356399536,
|
|
"epoch": 9.874432677760968,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 5.0188170827362294e-05,
|
|
"loss": 4.4296,
|
|
"mean_token_accuracy": 0.23001006841659546,
|
|
"num_tokens": 21636057.0,
|
|
"step": 9795
|
|
},
|
|
{
|
|
"entropy": 5.156159353256226,
|
|
"epoch": 9.879475542107917,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.017230083433317e-05,
|
|
"loss": 4.3975,
|
|
"mean_token_accuracy": 0.21633600294589997,
|
|
"num_tokens": 21646304.0,
|
|
"step": 9800
|
|
},
|
|
{
|
|
"entropy": 5.1467469215393065,
|
|
"epoch": 9.884518406454866,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.015712961097938e-05,
|
|
"loss": 4.3865,
|
|
"mean_token_accuracy": 0.23039887100458145,
|
|
"num_tokens": 21657867.0,
|
|
"step": 9805
|
|
},
|
|
{
|
|
"entropy": 5.1646452903747555,
|
|
"epoch": 9.889561270801815,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.0142657204453424e-05,
|
|
"loss": 4.4057,
|
|
"mean_token_accuracy": 0.23388028740882874,
|
|
"num_tokens": 21669017.0,
|
|
"step": 9810
|
|
},
|
|
{
|
|
"entropy": 5.128488492965698,
|
|
"epoch": 9.894604135148764,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 5.012888365973588e-05,
|
|
"loss": 4.4119,
|
|
"mean_token_accuracy": 0.22625089883804322,
|
|
"num_tokens": 21679712.0,
|
|
"step": 9815
|
|
},
|
|
{
|
|
"entropy": 5.153301477432251,
|
|
"epoch": 9.899646999495713,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 5.0115809019635217e-05,
|
|
"loss": 4.4104,
|
|
"mean_token_accuracy": 0.22174053639173508,
|
|
"num_tokens": 21692401.0,
|
|
"step": 9820
|
|
},
|
|
{
|
|
"entropy": 5.1751350402832035,
|
|
"epoch": 9.904689863842663,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 5.010343332478776e-05,
|
|
"loss": 4.4304,
|
|
"mean_token_accuracy": 0.2281507283449173,
|
|
"num_tokens": 21703748.0,
|
|
"step": 9825
|
|
},
|
|
{
|
|
"entropy": 5.152105760574341,
|
|
"epoch": 9.909732728189612,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 5.009175661365743e-05,
|
|
"loss": 4.4502,
|
|
"mean_token_accuracy": 0.2209807887673378,
|
|
"num_tokens": 21714363.0,
|
|
"step": 9830
|
|
},
|
|
{
|
|
"entropy": 5.125897073745728,
|
|
"epoch": 9.91477559253656,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 5.00807789225357e-05,
|
|
"loss": 4.4074,
|
|
"mean_token_accuracy": 0.23163869678974153,
|
|
"num_tokens": 21725818.0,
|
|
"step": 9835
|
|
},
|
|
{
|
|
"entropy": 5.12597246170044,
|
|
"epoch": 9.91981845688351,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 5.007050028554153e-05,
|
|
"loss": 4.3621,
|
|
"mean_token_accuracy": 0.2215610697865486,
|
|
"num_tokens": 21737009.0,
|
|
"step": 9840
|
|
},
|
|
{
|
|
"entropy": 5.077091264724731,
|
|
"epoch": 9.924861321230459,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.006092073462112e-05,
|
|
"loss": 4.3098,
|
|
"mean_token_accuracy": 0.23328551203012465,
|
|
"num_tokens": 21747301.0,
|
|
"step": 9845
|
|
},
|
|
{
|
|
"entropy": 5.162187194824218,
|
|
"epoch": 9.929904185577408,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.005204029954797e-05,
|
|
"loss": 4.3934,
|
|
"mean_token_accuracy": 0.2284982070326805,
|
|
"num_tokens": 21758236.0,
|
|
"step": 9850
|
|
},
|
|
{
|
|
"entropy": 5.144351577758789,
|
|
"epoch": 9.934947049924357,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.004385900792263e-05,
|
|
"loss": 4.399,
|
|
"mean_token_accuracy": 0.22760124057531356,
|
|
"num_tokens": 21768850.0,
|
|
"step": 9855
|
|
},
|
|
{
|
|
"entropy": 5.119643306732177,
|
|
"epoch": 9.939989914271306,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 5.003637688517278e-05,
|
|
"loss": 4.2693,
|
|
"mean_token_accuracy": 0.23518936336040497,
|
|
"num_tokens": 21778556.0,
|
|
"step": 9860
|
|
},
|
|
{
|
|
"entropy": 5.1891741275787355,
|
|
"epoch": 9.945032778618256,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 5.002959395455303e-05,
|
|
"loss": 4.3738,
|
|
"mean_token_accuracy": 0.23457685559988023,
|
|
"num_tokens": 21789814.0,
|
|
"step": 9865
|
|
},
|
|
{
|
|
"entropy": 5.2017737865448,
|
|
"epoch": 9.950075642965205,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.0023510237144885e-05,
|
|
"loss": 4.4071,
|
|
"mean_token_accuracy": 0.2276877298951149,
|
|
"num_tokens": 21800681.0,
|
|
"step": 9870
|
|
},
|
|
{
|
|
"entropy": 5.131738138198853,
|
|
"epoch": 9.955118507312154,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 5.001812575185665e-05,
|
|
"loss": 4.3851,
|
|
"mean_token_accuracy": 0.2297256514430046,
|
|
"num_tokens": 21814177.0,
|
|
"step": 9875
|
|
},
|
|
{
|
|
"entropy": 5.140628862380981,
|
|
"epoch": 9.960161371659103,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.001344051542347e-05,
|
|
"loss": 4.3569,
|
|
"mean_token_accuracy": 0.23275707215070723,
|
|
"num_tokens": 21826649.0,
|
|
"step": 9880
|
|
},
|
|
{
|
|
"entropy": 5.11033821105957,
|
|
"epoch": 9.965204236006052,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 5.0009454542407126e-05,
|
|
"loss": 4.3954,
|
|
"mean_token_accuracy": 0.22828697562217712,
|
|
"num_tokens": 21837912.0,
|
|
"step": 9885
|
|
},
|
|
{
|
|
"entropy": 5.168292617797851,
|
|
"epoch": 9.970247100353001,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.000616784519616e-05,
|
|
"loss": 4.3881,
|
|
"mean_token_accuracy": 0.2243731439113617,
|
|
"num_tokens": 21849139.0,
|
|
"step": 9890
|
|
},
|
|
{
|
|
"entropy": 5.1493370056152346,
|
|
"epoch": 9.97528996469995,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 5.000358043400566e-05,
|
|
"loss": 4.3823,
|
|
"mean_token_accuracy": 0.22565737068653108,
|
|
"num_tokens": 21859713.0,
|
|
"step": 9895
|
|
},
|
|
{
|
|
"entropy": 5.142309904098511,
|
|
"epoch": 9.9803328290469,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 5.000169231687739e-05,
|
|
"loss": 4.3324,
|
|
"mean_token_accuracy": 0.24203885197639466,
|
|
"num_tokens": 21871267.0,
|
|
"step": 9900
|
|
},
|
|
{
|
|
"entropy": 5.120758438110352,
|
|
"epoch": 9.985375693393848,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 5.000050349967963e-05,
|
|
"loss": 4.369,
|
|
"mean_token_accuracy": 0.22867218852043153,
|
|
"num_tokens": 21883014.0,
|
|
"step": 9905
|
|
},
|
|
{
|
|
"entropy": 5.11875057220459,
|
|
"epoch": 9.990418557740798,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 5.000001398610728e-05,
|
|
"loss": 4.3888,
|
|
"mean_token_accuracy": 0.2348761960864067,
|
|
"num_tokens": 21894532.0,
|
|
"step": 9910
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9910,
|
|
"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": true
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 3.6572881390848e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|