2475 lines
69 KiB
JSON
2475 lines
69 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 3.2,
|
||
|
|
"eval_steps": 100,
|
||
|
|
"global_step": 2000,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"entropy": 0.08180969459936023,
|
||
|
|
"epoch": 0.016,
|
||
|
|
"grad_norm": 0.67578125,
|
||
|
|
"learning_rate": 9.000000000000001e-07,
|
||
|
|
"loss": 0.0453,
|
||
|
|
"mean_token_accuracy": 0.9865816306322813,
|
||
|
|
"num_tokens": 195510.0,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08111731857061386,
|
||
|
|
"epoch": 0.032,
|
||
|
|
"grad_norm": 0.60546875,
|
||
|
|
"learning_rate": 1.9000000000000002e-06,
|
||
|
|
"loss": 0.0452,
|
||
|
|
"mean_token_accuracy": 0.986666041240096,
|
||
|
|
"num_tokens": 391008.0,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.07587501825764775,
|
||
|
|
"epoch": 0.048,
|
||
|
|
"grad_norm": 0.6953125,
|
||
|
|
"learning_rate": 2.9e-06,
|
||
|
|
"loss": 0.0445,
|
||
|
|
"mean_token_accuracy": 0.9873391918838024,
|
||
|
|
"num_tokens": 590136.0,
|
||
|
|
"step": 30
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.07685074284672737,
|
||
|
|
"epoch": 0.064,
|
||
|
|
"grad_norm": 0.625,
|
||
|
|
"learning_rate": 3.900000000000001e-06,
|
||
|
|
"loss": 0.0441,
|
||
|
|
"mean_token_accuracy": 0.987066001445055,
|
||
|
|
"num_tokens": 791432.0,
|
||
|
|
"step": 40
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.07894687210209668,
|
||
|
|
"epoch": 0.08,
|
||
|
|
"grad_norm": 0.75390625,
|
||
|
|
"learning_rate": 4.9000000000000005e-06,
|
||
|
|
"loss": 0.0447,
|
||
|
|
"mean_token_accuracy": 0.9865922331809998,
|
||
|
|
"num_tokens": 990342.0,
|
||
|
|
"step": 50
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.07919039381667972,
|
||
|
|
"epoch": 0.096,
|
||
|
|
"grad_norm": 0.66015625,
|
||
|
|
"learning_rate": 5.9e-06,
|
||
|
|
"loss": 0.0466,
|
||
|
|
"mean_token_accuracy": 0.9863619733601808,
|
||
|
|
"num_tokens": 1182339.0,
|
||
|
|
"step": 60
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.07967684981413185,
|
||
|
|
"epoch": 0.112,
|
||
|
|
"grad_norm": 0.5859375,
|
||
|
|
"learning_rate": 6.9e-06,
|
||
|
|
"loss": 0.0447,
|
||
|
|
"mean_token_accuracy": 0.9870311982929707,
|
||
|
|
"num_tokens": 1386193.0,
|
||
|
|
"step": 70
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08054462079890072,
|
||
|
|
"epoch": 0.128,
|
||
|
|
"grad_norm": 0.65234375,
|
||
|
|
"learning_rate": 7.9e-06,
|
||
|
|
"loss": 0.0466,
|
||
|
|
"mean_token_accuracy": 0.9863289542496204,
|
||
|
|
"num_tokens": 1583261.0,
|
||
|
|
"step": 80
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08421374012250453,
|
||
|
|
"epoch": 0.144,
|
||
|
|
"grad_norm": 0.5546875,
|
||
|
|
"learning_rate": 8.900000000000001e-06,
|
||
|
|
"loss": 0.0489,
|
||
|
|
"mean_token_accuracy": 0.9860724166035653,
|
||
|
|
"num_tokens": 1774826.0,
|
||
|
|
"step": 90
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08307966932188719,
|
||
|
|
"epoch": 0.16,
|
||
|
|
"grad_norm": 0.71484375,
|
||
|
|
"learning_rate": 9.9e-06,
|
||
|
|
"loss": 0.0489,
|
||
|
|
"mean_token_accuracy": 0.986110083758831,
|
||
|
|
"num_tokens": 1971310.0,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.16,
|
||
|
|
"eval_biology_entropy": 0.3792700763940811,
|
||
|
|
"eval_biology_loss": 2.260571241378784,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.6872955484390258,
|
||
|
|
"eval_biology_num_tokens": 1971310.0,
|
||
|
|
"eval_biology_runtime": 39.5255,
|
||
|
|
"eval_biology_samples_per_second": 12.65,
|
||
|
|
"eval_biology_steps_per_second": 3.163,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.16,
|
||
|
|
"eval_chemistry_entropy": 0.23563545274734496,
|
||
|
|
"eval_chemistry_loss": 1.4561725854873657,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7805930047035218,
|
||
|
|
"eval_chemistry_num_tokens": 1971310.0,
|
||
|
|
"eval_chemistry_runtime": 49.2319,
|
||
|
|
"eval_chemistry_samples_per_second": 10.156,
|
||
|
|
"eval_chemistry_steps_per_second": 2.539,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08481427943333983,
|
||
|
|
"epoch": 0.176,
|
||
|
|
"grad_norm": 0.734375,
|
||
|
|
"learning_rate": 1.0900000000000002e-05,
|
||
|
|
"loss": 0.0497,
|
||
|
|
"mean_token_accuracy": 0.9861054014414549,
|
||
|
|
"num_tokens": 2169453.0,
|
||
|
|
"step": 110
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.0882645736914128,
|
||
|
|
"epoch": 0.192,
|
||
|
|
"grad_norm": 0.79296875,
|
||
|
|
"learning_rate": 1.1900000000000001e-05,
|
||
|
|
"loss": 0.0512,
|
||
|
|
"mean_token_accuracy": 0.9859256945550442,
|
||
|
|
"num_tokens": 2366844.0,
|
||
|
|
"step": 120
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.09435517250094563,
|
||
|
|
"epoch": 0.208,
|
||
|
|
"grad_norm": 1.2421875,
|
||
|
|
"learning_rate": 1.2900000000000002e-05,
|
||
|
|
"loss": 0.0547,
|
||
|
|
"mean_token_accuracy": 0.9849376413971186,
|
||
|
|
"num_tokens": 2559708.0,
|
||
|
|
"step": 130
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.09682020233012736,
|
||
|
|
"epoch": 0.224,
|
||
|
|
"grad_norm": 1.3671875,
|
||
|
|
"learning_rate": 1.39e-05,
|
||
|
|
"loss": 0.057,
|
||
|
|
"mean_token_accuracy": 0.985217172652483,
|
||
|
|
"num_tokens": 2756533.0,
|
||
|
|
"step": 140
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11111663887277246,
|
||
|
|
"epoch": 0.24,
|
||
|
|
"grad_norm": 2.84375,
|
||
|
|
"learning_rate": 1.4900000000000001e-05,
|
||
|
|
"loss": 0.0657,
|
||
|
|
"mean_token_accuracy": 0.9836616575717926,
|
||
|
|
"num_tokens": 2948600.0,
|
||
|
|
"step": 150
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12251729485578836,
|
||
|
|
"epoch": 0.256,
|
||
|
|
"grad_norm": 2.453125,
|
||
|
|
"learning_rate": 1.5900000000000004e-05,
|
||
|
|
"loss": 0.0728,
|
||
|
|
"mean_token_accuracy": 0.9814188834279776,
|
||
|
|
"num_tokens": 3141473.0,
|
||
|
|
"step": 160
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12813111576251685,
|
||
|
|
"epoch": 0.272,
|
||
|
|
"grad_norm": 3.03125,
|
||
|
|
"learning_rate": 1.69e-05,
|
||
|
|
"loss": 0.0758,
|
||
|
|
"mean_token_accuracy": 0.9808279071003199,
|
||
|
|
"num_tokens": 3337285.0,
|
||
|
|
"step": 170
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.13337696841917931,
|
||
|
|
"epoch": 0.288,
|
||
|
|
"grad_norm": 2.984375,
|
||
|
|
"learning_rate": 1.79e-05,
|
||
|
|
"loss": 0.0771,
|
||
|
|
"mean_token_accuracy": 0.9798636745661498,
|
||
|
|
"num_tokens": 3541221.0,
|
||
|
|
"step": 180
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.1436500378418714,
|
||
|
|
"epoch": 0.304,
|
||
|
|
"grad_norm": 2.78125,
|
||
|
|
"learning_rate": 1.8900000000000002e-05,
|
||
|
|
"loss": 0.0851,
|
||
|
|
"mean_token_accuracy": 0.9780430492013693,
|
||
|
|
"num_tokens": 3734952.0,
|
||
|
|
"step": 190
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.1549748638179153,
|
||
|
|
"epoch": 0.32,
|
||
|
|
"grad_norm": 3.390625,
|
||
|
|
"learning_rate": 1.9900000000000003e-05,
|
||
|
|
"loss": 0.0946,
|
||
|
|
"mean_token_accuracy": 0.9750579129904509,
|
||
|
|
"num_tokens": 3922209.0,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.32,
|
||
|
|
"eval_biology_entropy": 0.4169517788887024,
|
||
|
|
"eval_biology_loss": 2.160527229309082,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.6808431892395019,
|
||
|
|
"eval_biology_num_tokens": 3922209.0,
|
||
|
|
"eval_biology_runtime": 39.5121,
|
||
|
|
"eval_biology_samples_per_second": 12.654,
|
||
|
|
"eval_biology_steps_per_second": 3.164,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.32,
|
||
|
|
"eval_chemistry_entropy": 0.25474807441234587,
|
||
|
|
"eval_chemistry_loss": 1.3652101755142212,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7794562330245972,
|
||
|
|
"eval_chemistry_num_tokens": 3922209.0,
|
||
|
|
"eval_chemistry_runtime": 49.2263,
|
||
|
|
"eval_chemistry_samples_per_second": 10.157,
|
||
|
|
"eval_chemistry_steps_per_second": 2.539,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.16029730178415774,
|
||
|
|
"epoch": 0.336,
|
||
|
|
"grad_norm": 3.515625,
|
||
|
|
"learning_rate": 1.9900000000000003e-05,
|
||
|
|
"loss": 0.097,
|
||
|
|
"mean_token_accuracy": 0.9749548710882664,
|
||
|
|
"num_tokens": 4115941.0,
|
||
|
|
"step": 210
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.16590779088437557,
|
||
|
|
"epoch": 0.352,
|
||
|
|
"grad_norm": 3.78125,
|
||
|
|
"learning_rate": 1.978888888888889e-05,
|
||
|
|
"loss": 0.1023,
|
||
|
|
"mean_token_accuracy": 0.9726595059037209,
|
||
|
|
"num_tokens": 4308755.0,
|
||
|
|
"step": 220
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.1707212173845619,
|
||
|
|
"epoch": 0.368,
|
||
|
|
"grad_norm": 3.390625,
|
||
|
|
"learning_rate": 1.967777777777778e-05,
|
||
|
|
"loss": 0.1023,
|
||
|
|
"mean_token_accuracy": 0.9726378936320543,
|
||
|
|
"num_tokens": 4505910.0,
|
||
|
|
"step": 230
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.1765518521424383,
|
||
|
|
"epoch": 0.384,
|
||
|
|
"grad_norm": 3.671875,
|
||
|
|
"learning_rate": 1.956666666666667e-05,
|
||
|
|
"loss": 0.1083,
|
||
|
|
"mean_token_accuracy": 0.97143131531775,
|
||
|
|
"num_tokens": 4696101.0,
|
||
|
|
"step": 240
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.17895571915432812,
|
||
|
|
"epoch": 0.4,
|
||
|
|
"grad_norm": 3.5,
|
||
|
|
"learning_rate": 1.9455555555555555e-05,
|
||
|
|
"loss": 0.11,
|
||
|
|
"mean_token_accuracy": 0.9705049484968186,
|
||
|
|
"num_tokens": 4889445.0,
|
||
|
|
"step": 250
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.18151565054431557,
|
||
|
|
"epoch": 0.416,
|
||
|
|
"grad_norm": 3.578125,
|
||
|
|
"learning_rate": 1.9344444444444446e-05,
|
||
|
|
"loss": 0.1091,
|
||
|
|
"mean_token_accuracy": 0.970859233289957,
|
||
|
|
"num_tokens": 5087624.0,
|
||
|
|
"step": 260
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.18663504091091454,
|
||
|
|
"epoch": 0.432,
|
||
|
|
"grad_norm": 3.859375,
|
||
|
|
"learning_rate": 1.9233333333333334e-05,
|
||
|
|
"loss": 0.1135,
|
||
|
|
"mean_token_accuracy": 0.9693388067185879,
|
||
|
|
"num_tokens": 5273794.0,
|
||
|
|
"step": 270
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.18451756797730923,
|
||
|
|
"epoch": 0.448,
|
||
|
|
"grad_norm": 4.125,
|
||
|
|
"learning_rate": 1.9122222222222226e-05,
|
||
|
|
"loss": 0.1144,
|
||
|
|
"mean_token_accuracy": 0.9698665160685778,
|
||
|
|
"num_tokens": 5463288.0,
|
||
|
|
"step": 280
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.17932639280334114,
|
||
|
|
"epoch": 0.464,
|
||
|
|
"grad_norm": 3.5625,
|
||
|
|
"learning_rate": 1.9011111111111114e-05,
|
||
|
|
"loss": 0.1081,
|
||
|
|
"mean_token_accuracy": 0.9717006407678127,
|
||
|
|
"num_tokens": 5656827.0,
|
||
|
|
"step": 290
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.18305762661620975,
|
||
|
|
"epoch": 0.48,
|
||
|
|
"grad_norm": 3.90625,
|
||
|
|
"learning_rate": 1.8900000000000002e-05,
|
||
|
|
"loss": 0.1107,
|
||
|
|
"mean_token_accuracy": 0.9712053131312132,
|
||
|
|
"num_tokens": 5853311.0,
|
||
|
|
"step": 300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.48,
|
||
|
|
"eval_biology_entropy": 0.42041616535186765,
|
||
|
|
"eval_biology_loss": 2.1313436031341553,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.6798483271598816,
|
||
|
|
"eval_biology_num_tokens": 5853311.0,
|
||
|
|
"eval_biology_runtime": 39.5137,
|
||
|
|
"eval_biology_samples_per_second": 12.654,
|
||
|
|
"eval_biology_steps_per_second": 3.163,
|
||
|
|
"step": 300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.48,
|
||
|
|
"eval_chemistry_entropy": 0.259404793381691,
|
||
|
|
"eval_chemistry_loss": 1.3463311195373535,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7787248449325561,
|
||
|
|
"eval_chemistry_num_tokens": 5853311.0,
|
||
|
|
"eval_chemistry_runtime": 49.201,
|
||
|
|
"eval_chemistry_samples_per_second": 10.162,
|
||
|
|
"eval_chemistry_steps_per_second": 2.541,
|
||
|
|
"step": 300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.17875824291259051,
|
||
|
|
"epoch": 0.496,
|
||
|
|
"grad_norm": 3.453125,
|
||
|
|
"learning_rate": 1.878888888888889e-05,
|
||
|
|
"loss": 0.1101,
|
||
|
|
"mean_token_accuracy": 0.9710387323051691,
|
||
|
|
"num_tokens": 6049788.0,
|
||
|
|
"step": 310
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.1863187742419541,
|
||
|
|
"epoch": 0.512,
|
||
|
|
"grad_norm": 3.4375,
|
||
|
|
"learning_rate": 1.8677777777777778e-05,
|
||
|
|
"loss": 0.113,
|
||
|
|
"mean_token_accuracy": 0.9698710650205612,
|
||
|
|
"num_tokens": 6243825.0,
|
||
|
|
"step": 320
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.18619006094522775,
|
||
|
|
"epoch": 0.528,
|
||
|
|
"grad_norm": 3.890625,
|
||
|
|
"learning_rate": 1.856666666666667e-05,
|
||
|
|
"loss": 0.1135,
|
||
|
|
"mean_token_accuracy": 0.9703223951160908,
|
||
|
|
"num_tokens": 6434188.0,
|
||
|
|
"step": 330
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.18118073977530003,
|
||
|
|
"epoch": 0.544,
|
||
|
|
"grad_norm": 3.484375,
|
||
|
|
"learning_rate": 1.8455555555555558e-05,
|
||
|
|
"loss": 0.1114,
|
||
|
|
"mean_token_accuracy": 0.9708513412624598,
|
||
|
|
"num_tokens": 6629832.0,
|
||
|
|
"step": 340
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.1808169964235276,
|
||
|
|
"epoch": 0.56,
|
||
|
|
"grad_norm": 3.28125,
|
||
|
|
"learning_rate": 1.8344444444444446e-05,
|
||
|
|
"loss": 0.1079,
|
||
|
|
"mean_token_accuracy": 0.9716686312109232,
|
||
|
|
"num_tokens": 6824668.0,
|
||
|
|
"step": 350
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.18033515606075526,
|
||
|
|
"epoch": 0.576,
|
||
|
|
"grad_norm": 3.53125,
|
||
|
|
"learning_rate": 1.8233333333333334e-05,
|
||
|
|
"loss": 0.106,
|
||
|
|
"mean_token_accuracy": 0.9722726982086897,
|
||
|
|
"num_tokens": 7025923.0,
|
||
|
|
"step": 360
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.1852203646209091,
|
||
|
|
"epoch": 0.592,
|
||
|
|
"grad_norm": 3.625,
|
||
|
|
"learning_rate": 1.8122222222222222e-05,
|
||
|
|
"loss": 0.111,
|
||
|
|
"mean_token_accuracy": 0.9705461781471968,
|
||
|
|
"num_tokens": 7218224.0,
|
||
|
|
"step": 370
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.17758405841886998,
|
||
|
|
"epoch": 0.608,
|
||
|
|
"grad_norm": 3.65625,
|
||
|
|
"learning_rate": 1.8011111111111113e-05,
|
||
|
|
"loss": 0.1066,
|
||
|
|
"mean_token_accuracy": 0.9723859451711178,
|
||
|
|
"num_tokens": 7421216.0,
|
||
|
|
"step": 380
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.18370326030999423,
|
||
|
|
"epoch": 0.624,
|
||
|
|
"grad_norm": 3.765625,
|
||
|
|
"learning_rate": 1.79e-05,
|
||
|
|
"loss": 0.1096,
|
||
|
|
"mean_token_accuracy": 0.9708739675581455,
|
||
|
|
"num_tokens": 7617258.0,
|
||
|
|
"step": 390
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.1831163068767637,
|
||
|
|
"epoch": 0.64,
|
||
|
|
"grad_norm": 3.578125,
|
||
|
|
"learning_rate": 1.778888888888889e-05,
|
||
|
|
"loss": 0.1072,
|
||
|
|
"mean_token_accuracy": 0.9717655830085278,
|
||
|
|
"num_tokens": 7806173.0,
|
||
|
|
"step": 400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.64,
|
||
|
|
"eval_biology_entropy": 0.42119112753868104,
|
||
|
|
"eval_biology_loss": 2.1267058849334717,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.6801503438949585,
|
||
|
|
"eval_biology_num_tokens": 7806173.0,
|
||
|
|
"eval_biology_runtime": 39.5121,
|
||
|
|
"eval_biology_samples_per_second": 12.654,
|
||
|
|
"eval_biology_steps_per_second": 3.164,
|
||
|
|
"step": 400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.64,
|
||
|
|
"eval_chemistry_entropy": 0.2595027017593384,
|
||
|
|
"eval_chemistry_loss": 1.3404533863067627,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7790702543258667,
|
||
|
|
"eval_chemistry_num_tokens": 7806173.0,
|
||
|
|
"eval_chemistry_runtime": 49.2002,
|
||
|
|
"eval_chemistry_samples_per_second": 10.163,
|
||
|
|
"eval_chemistry_steps_per_second": 2.541,
|
||
|
|
"step": 400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.1827191138640046,
|
||
|
|
"epoch": 0.656,
|
||
|
|
"grad_norm": 3.84375,
|
||
|
|
"learning_rate": 1.7677777777777778e-05,
|
||
|
|
"loss": 0.1088,
|
||
|
|
"mean_token_accuracy": 0.9715055178850889,
|
||
|
|
"num_tokens": 8000514.0,
|
||
|
|
"step": 410
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.18661352493800223,
|
||
|
|
"epoch": 0.672,
|
||
|
|
"grad_norm": 3.796875,
|
||
|
|
"learning_rate": 1.756666666666667e-05,
|
||
|
|
"loss": 0.1114,
|
||
|
|
"mean_token_accuracy": 0.9704575203359127,
|
||
|
|
"num_tokens": 8187884.0,
|
||
|
|
"step": 420
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.17729734890162946,
|
||
|
|
"epoch": 0.688,
|
||
|
|
"grad_norm": 3.09375,
|
||
|
|
"learning_rate": 1.7455555555555557e-05,
|
||
|
|
"loss": 0.1046,
|
||
|
|
"mean_token_accuracy": 0.9725840430706739,
|
||
|
|
"num_tokens": 8390643.0,
|
||
|
|
"step": 430
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.18038612157106398,
|
||
|
|
"epoch": 0.704,
|
||
|
|
"grad_norm": 3.46875,
|
||
|
|
"learning_rate": 1.7344444444444445e-05,
|
||
|
|
"loss": 0.1072,
|
||
|
|
"mean_token_accuracy": 0.9721495136618614,
|
||
|
|
"num_tokens": 8583266.0,
|
||
|
|
"step": 440
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.17609651018865408,
|
||
|
|
"epoch": 0.72,
|
||
|
|
"grad_norm": 3.734375,
|
||
|
|
"learning_rate": 1.7233333333333337e-05,
|
||
|
|
"loss": 0.1045,
|
||
|
|
"mean_token_accuracy": 0.9718786559998989,
|
||
|
|
"num_tokens": 8786233.0,
|
||
|
|
"step": 450
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.1797325788065791,
|
||
|
|
"epoch": 0.736,
|
||
|
|
"grad_norm": 3.390625,
|
||
|
|
"learning_rate": 1.7122222222222225e-05,
|
||
|
|
"loss": 0.107,
|
||
|
|
"mean_token_accuracy": 0.9716725051403046,
|
||
|
|
"num_tokens": 8982898.0,
|
||
|
|
"step": 460
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.17950535458512604,
|
||
|
|
"epoch": 0.752,
|
||
|
|
"grad_norm": 3.5625,
|
||
|
|
"learning_rate": 1.7011111111111113e-05,
|
||
|
|
"loss": 0.1072,
|
||
|
|
"mean_token_accuracy": 0.9725060146301985,
|
||
|
|
"num_tokens": 9174361.0,
|
||
|
|
"step": 470
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.17792485035024583,
|
||
|
|
"epoch": 0.768,
|
||
|
|
"grad_norm": 3.375,
|
||
|
|
"learning_rate": 1.69e-05,
|
||
|
|
"loss": 0.1025,
|
||
|
|
"mean_token_accuracy": 0.973187630996108,
|
||
|
|
"num_tokens": 9373430.0,
|
||
|
|
"step": 480
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.17844869694672524,
|
||
|
|
"epoch": 0.784,
|
||
|
|
"grad_norm": 3.515625,
|
||
|
|
"learning_rate": 1.678888888888889e-05,
|
||
|
|
"loss": 0.1073,
|
||
|
|
"mean_token_accuracy": 0.9717021223157645,
|
||
|
|
"num_tokens": 9570409.0,
|
||
|
|
"step": 490
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.17820251639932394,
|
||
|
|
"epoch": 0.8,
|
||
|
|
"grad_norm": 3.765625,
|
||
|
|
"learning_rate": 1.667777777777778e-05,
|
||
|
|
"loss": 0.1032,
|
||
|
|
"mean_token_accuracy": 0.9729967709630728,
|
||
|
|
"num_tokens": 9766360.0,
|
||
|
|
"step": 500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8,
|
||
|
|
"eval_biology_entropy": 0.41608734560012817,
|
||
|
|
"eval_biology_loss": 2.139784097671509,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.6806383662223816,
|
||
|
|
"eval_biology_num_tokens": 9766360.0,
|
||
|
|
"eval_biology_runtime": 39.5271,
|
||
|
|
"eval_biology_samples_per_second": 12.65,
|
||
|
|
"eval_biology_steps_per_second": 3.162,
|
||
|
|
"step": 500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8,
|
||
|
|
"eval_chemistry_entropy": 0.25711704087257387,
|
||
|
|
"eval_chemistry_loss": 1.3562349081039429,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7787247653007507,
|
||
|
|
"eval_chemistry_num_tokens": 9766360.0,
|
||
|
|
"eval_chemistry_runtime": 48.8194,
|
||
|
|
"eval_chemistry_samples_per_second": 10.242,
|
||
|
|
"eval_chemistry_steps_per_second": 2.56,
|
||
|
|
"step": 500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.1748917239718139,
|
||
|
|
"epoch": 0.816,
|
||
|
|
"grad_norm": 3.515625,
|
||
|
|
"learning_rate": 1.656666666666667e-05,
|
||
|
|
"loss": 0.1004,
|
||
|
|
"mean_token_accuracy": 0.9740771554410458,
|
||
|
|
"num_tokens": 9963834.0,
|
||
|
|
"step": 510
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.17268717400729655,
|
||
|
|
"epoch": 0.832,
|
||
|
|
"grad_norm": 3.125,
|
||
|
|
"learning_rate": 1.6455555555555556e-05,
|
||
|
|
"loss": 0.1023,
|
||
|
|
"mean_token_accuracy": 0.9740318663418293,
|
||
|
|
"num_tokens": 10160950.0,
|
||
|
|
"step": 520
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.17282785503193737,
|
||
|
|
"epoch": 0.848,
|
||
|
|
"grad_norm": 3.671875,
|
||
|
|
"learning_rate": 1.6344444444444445e-05,
|
||
|
|
"loss": 0.0991,
|
||
|
|
"mean_token_accuracy": 0.9744720965623855,
|
||
|
|
"num_tokens": 10363131.0,
|
||
|
|
"step": 530
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.17644098605960606,
|
||
|
|
"epoch": 0.864,
|
||
|
|
"grad_norm": 3.34375,
|
||
|
|
"learning_rate": 1.6233333333333333e-05,
|
||
|
|
"loss": 0.1026,
|
||
|
|
"mean_token_accuracy": 0.972842201218009,
|
||
|
|
"num_tokens": 10558003.0,
|
||
|
|
"step": 540
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.17435273211449384,
|
||
|
|
"epoch": 0.88,
|
||
|
|
"grad_norm": 3.34375,
|
||
|
|
"learning_rate": 1.6122222222222224e-05,
|
||
|
|
"loss": 0.1004,
|
||
|
|
"mean_token_accuracy": 0.9740745916962623,
|
||
|
|
"num_tokens": 10754470.0,
|
||
|
|
"step": 550
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.1750097878277302,
|
||
|
|
"epoch": 0.896,
|
||
|
|
"grad_norm": 3.453125,
|
||
|
|
"learning_rate": 1.6011111111111112e-05,
|
||
|
|
"loss": 0.1007,
|
||
|
|
"mean_token_accuracy": 0.9743314851075411,
|
||
|
|
"num_tokens": 10948943.0,
|
||
|
|
"step": 560
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.17419813955202698,
|
||
|
|
"epoch": 0.912,
|
||
|
|
"grad_norm": 3.234375,
|
||
|
|
"learning_rate": 1.5900000000000004e-05,
|
||
|
|
"loss": 0.1005,
|
||
|
|
"mean_token_accuracy": 0.974232492223382,
|
||
|
|
"num_tokens": 11142613.0,
|
||
|
|
"step": 570
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.17595540760084988,
|
||
|
|
"epoch": 0.928,
|
||
|
|
"grad_norm": 3.546875,
|
||
|
|
"learning_rate": 1.5788888888888888e-05,
|
||
|
|
"loss": 0.0997,
|
||
|
|
"mean_token_accuracy": 0.974004115909338,
|
||
|
|
"num_tokens": 11336779.0,
|
||
|
|
"step": 580
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.16677158032543957,
|
||
|
|
"epoch": 0.944,
|
||
|
|
"grad_norm": 3.140625,
|
||
|
|
"learning_rate": 1.567777777777778e-05,
|
||
|
|
"loss": 0.0982,
|
||
|
|
"mean_token_accuracy": 0.9750826183706522,
|
||
|
|
"num_tokens": 11536381.0,
|
||
|
|
"step": 590
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.16921130288392305,
|
||
|
|
"epoch": 0.96,
|
||
|
|
"grad_norm": 3.546875,
|
||
|
|
"learning_rate": 1.5566666666666668e-05,
|
||
|
|
"loss": 0.0966,
|
||
|
|
"mean_token_accuracy": 0.9753067292273044,
|
||
|
|
"num_tokens": 11735527.0,
|
||
|
|
"step": 600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.96,
|
||
|
|
"eval_biology_entropy": 0.4160035889148712,
|
||
|
|
"eval_biology_loss": 2.1477200984954834,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.68089803981781,
|
||
|
|
"eval_biology_num_tokens": 11735527.0,
|
||
|
|
"eval_biology_runtime": 39.5581,
|
||
|
|
"eval_biology_samples_per_second": 12.64,
|
||
|
|
"eval_biology_steps_per_second": 3.16,
|
||
|
|
"step": 600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.96,
|
||
|
|
"eval_chemistry_entropy": 0.2561914825439453,
|
||
|
|
"eval_chemistry_loss": 1.3602544069290161,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.779716450214386,
|
||
|
|
"eval_chemistry_num_tokens": 11735527.0,
|
||
|
|
"eval_chemistry_runtime": 49.254,
|
||
|
|
"eval_chemistry_samples_per_second": 10.151,
|
||
|
|
"eval_chemistry_steps_per_second": 2.538,
|
||
|
|
"step": 600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.16858043959364294,
|
||
|
|
"epoch": 0.976,
|
||
|
|
"grad_norm": 3.65625,
|
||
|
|
"learning_rate": 1.5455555555555556e-05,
|
||
|
|
"loss": 0.0985,
|
||
|
|
"mean_token_accuracy": 0.9748617243021727,
|
||
|
|
"num_tokens": 11930935.0,
|
||
|
|
"step": 610
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.16747406986542046,
|
||
|
|
"epoch": 0.992,
|
||
|
|
"grad_norm": 3.15625,
|
||
|
|
"learning_rate": 1.5344444444444447e-05,
|
||
|
|
"loss": 0.0955,
|
||
|
|
"mean_token_accuracy": 0.9756584648042917,
|
||
|
|
"num_tokens": 12129585.0,
|
||
|
|
"step": 620
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.15082475617527963,
|
||
|
|
"epoch": 1.008,
|
||
|
|
"grad_norm": 2.28125,
|
||
|
|
"learning_rate": 1.5233333333333335e-05,
|
||
|
|
"loss": 0.0856,
|
||
|
|
"mean_token_accuracy": 0.9783346977084875,
|
||
|
|
"num_tokens": 12325847.0,
|
||
|
|
"step": 630
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11756393420509995,
|
||
|
|
"epoch": 1.024,
|
||
|
|
"grad_norm": 2.3125,
|
||
|
|
"learning_rate": 1.5122222222222222e-05,
|
||
|
|
"loss": 0.07,
|
||
|
|
"mean_token_accuracy": 0.9816608231514692,
|
||
|
|
"num_tokens": 12530820.0,
|
||
|
|
"step": 640
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12562460631597788,
|
||
|
|
"epoch": 1.04,
|
||
|
|
"grad_norm": 2.796875,
|
||
|
|
"learning_rate": 1.5011111111111112e-05,
|
||
|
|
"loss": 0.0744,
|
||
|
|
"mean_token_accuracy": 0.9802271891385317,
|
||
|
|
"num_tokens": 12725384.0,
|
||
|
|
"step": 650
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12352396142669023,
|
||
|
|
"epoch": 1.056,
|
||
|
|
"grad_norm": 2.609375,
|
||
|
|
"learning_rate": 1.4900000000000001e-05,
|
||
|
|
"loss": 0.0744,
|
||
|
|
"mean_token_accuracy": 0.9802641972899437,
|
||
|
|
"num_tokens": 12924592.0,
|
||
|
|
"step": 660
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12865133616141974,
|
||
|
|
"epoch": 1.072,
|
||
|
|
"grad_norm": 2.671875,
|
||
|
|
"learning_rate": 1.4788888888888891e-05,
|
||
|
|
"loss": 0.0774,
|
||
|
|
"mean_token_accuracy": 0.9798023611307144,
|
||
|
|
"num_tokens": 13112848.0,
|
||
|
|
"step": 670
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12567441049031913,
|
||
|
|
"epoch": 1.088,
|
||
|
|
"grad_norm": 2.53125,
|
||
|
|
"learning_rate": 1.467777777777778e-05,
|
||
|
|
"loss": 0.0756,
|
||
|
|
"mean_token_accuracy": 0.9806226130574942,
|
||
|
|
"num_tokens": 13305725.0,
|
||
|
|
"step": 680
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12565653487108647,
|
||
|
|
"epoch": 1.104,
|
||
|
|
"grad_norm": 2.234375,
|
||
|
|
"learning_rate": 1.4566666666666669e-05,
|
||
|
|
"loss": 0.0745,
|
||
|
|
"mean_token_accuracy": 0.980208383128047,
|
||
|
|
"num_tokens": 13498685.0,
|
||
|
|
"step": 690
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.1318228157237172,
|
||
|
|
"epoch": 1.12,
|
||
|
|
"grad_norm": 2.5,
|
||
|
|
"learning_rate": 1.4455555555555555e-05,
|
||
|
|
"loss": 0.0778,
|
||
|
|
"mean_token_accuracy": 0.97950509339571,
|
||
|
|
"num_tokens": 13682146.0,
|
||
|
|
"step": 700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.12,
|
||
|
|
"eval_biology_entropy": 0.4004362862110138,
|
||
|
|
"eval_biology_loss": 2.177130699157715,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.6821538486480713,
|
||
|
|
"eval_biology_num_tokens": 13682146.0,
|
||
|
|
"eval_biology_runtime": 39.5089,
|
||
|
|
"eval_biology_samples_per_second": 12.655,
|
||
|
|
"eval_biology_steps_per_second": 3.164,
|
||
|
|
"step": 700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.12,
|
||
|
|
"eval_chemistry_entropy": 0.24954386246204377,
|
||
|
|
"eval_chemistry_loss": 1.3777952194213867,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7797037825584412,
|
||
|
|
"eval_chemistry_num_tokens": 13682146.0,
|
||
|
|
"eval_chemistry_runtime": 49.2017,
|
||
|
|
"eval_chemistry_samples_per_second": 10.162,
|
||
|
|
"eval_chemistry_steps_per_second": 2.541,
|
||
|
|
"step": 700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12691126235295086,
|
||
|
|
"epoch": 1.1360000000000001,
|
||
|
|
"grad_norm": 2.6875,
|
||
|
|
"learning_rate": 1.4344444444444445e-05,
|
||
|
|
"loss": 0.0757,
|
||
|
|
"mean_token_accuracy": 0.980040468648076,
|
||
|
|
"num_tokens": 13876656.0,
|
||
|
|
"step": 710
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12381730475462974,
|
||
|
|
"epoch": 1.152,
|
||
|
|
"grad_norm": 2.359375,
|
||
|
|
"learning_rate": 1.4233333333333335e-05,
|
||
|
|
"loss": 0.0743,
|
||
|
|
"mean_token_accuracy": 0.980041703954339,
|
||
|
|
"num_tokens": 14086104.0,
|
||
|
|
"step": 720
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.1285357222892344,
|
||
|
|
"epoch": 1.168,
|
||
|
|
"grad_norm": 2.5625,
|
||
|
|
"learning_rate": 1.4122222222222225e-05,
|
||
|
|
"loss": 0.0763,
|
||
|
|
"mean_token_accuracy": 0.9799800109118223,
|
||
|
|
"num_tokens": 14274520.0,
|
||
|
|
"step": 730
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12321301018819213,
|
||
|
|
"epoch": 1.184,
|
||
|
|
"grad_norm": 2.375,
|
||
|
|
"learning_rate": 1.4011111111111113e-05,
|
||
|
|
"loss": 0.0734,
|
||
|
|
"mean_token_accuracy": 0.9810513865202666,
|
||
|
|
"num_tokens": 14473623.0,
|
||
|
|
"step": 740
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.1292376267723739,
|
||
|
|
"epoch": 1.2,
|
||
|
|
"grad_norm": 2.609375,
|
||
|
|
"learning_rate": 1.39e-05,
|
||
|
|
"loss": 0.0784,
|
||
|
|
"mean_token_accuracy": 0.9795571226626635,
|
||
|
|
"num_tokens": 14661238.0,
|
||
|
|
"step": 750
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12086025532335043,
|
||
|
|
"epoch": 1.216,
|
||
|
|
"grad_norm": 2.640625,
|
||
|
|
"learning_rate": 1.3788888888888889e-05,
|
||
|
|
"loss": 0.073,
|
||
|
|
"mean_token_accuracy": 0.9810031026601791,
|
||
|
|
"num_tokens": 14865915.0,
|
||
|
|
"step": 760
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.1262558945454657,
|
||
|
|
"epoch": 1.232,
|
||
|
|
"grad_norm": 2.40625,
|
||
|
|
"learning_rate": 1.3677777777777779e-05,
|
||
|
|
"loss": 0.0786,
|
||
|
|
"mean_token_accuracy": 0.9797264195978641,
|
||
|
|
"num_tokens": 15055682.0,
|
||
|
|
"step": 770
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12587894778698683,
|
||
|
|
"epoch": 1.248,
|
||
|
|
"grad_norm": 2.484375,
|
||
|
|
"learning_rate": 1.3566666666666668e-05,
|
||
|
|
"loss": 0.0744,
|
||
|
|
"mean_token_accuracy": 0.9802309483289718,
|
||
|
|
"num_tokens": 15249576.0,
|
||
|
|
"step": 780
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12361958674155175,
|
||
|
|
"epoch": 1.264,
|
||
|
|
"grad_norm": 2.1875,
|
||
|
|
"learning_rate": 1.3455555555555556e-05,
|
||
|
|
"loss": 0.0726,
|
||
|
|
"mean_token_accuracy": 0.9805784326046705,
|
||
|
|
"num_tokens": 15446149.0,
|
||
|
|
"step": 790
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12549991067498922,
|
||
|
|
"epoch": 1.28,
|
||
|
|
"grad_norm": 2.546875,
|
||
|
|
"learning_rate": 1.3344444444444446e-05,
|
||
|
|
"loss": 0.0743,
|
||
|
|
"mean_token_accuracy": 0.9805208504199981,
|
||
|
|
"num_tokens": 15639317.0,
|
||
|
|
"step": 800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.28,
|
||
|
|
"eval_biology_entropy": 0.4032972083091736,
|
||
|
|
"eval_biology_loss": 2.1712591648101807,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.6816940937042236,
|
||
|
|
"eval_biology_num_tokens": 15639317.0,
|
||
|
|
"eval_biology_runtime": 39.5209,
|
||
|
|
"eval_biology_samples_per_second": 12.652,
|
||
|
|
"eval_biology_steps_per_second": 3.163,
|
||
|
|
"step": 800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.28,
|
||
|
|
"eval_chemistry_entropy": 0.25105089938640596,
|
||
|
|
"eval_chemistry_loss": 1.3797656297683716,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7792046613693238,
|
||
|
|
"eval_chemistry_num_tokens": 15639317.0,
|
||
|
|
"eval_chemistry_runtime": 49.2346,
|
||
|
|
"eval_chemistry_samples_per_second": 10.155,
|
||
|
|
"eval_chemistry_steps_per_second": 2.539,
|
||
|
|
"step": 800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12616121221799403,
|
||
|
|
"epoch": 1.296,
|
||
|
|
"grad_norm": 2.203125,
|
||
|
|
"learning_rate": 1.3233333333333334e-05,
|
||
|
|
"loss": 0.075,
|
||
|
|
"mean_token_accuracy": 0.9805853247642518,
|
||
|
|
"num_tokens": 15835805.0,
|
||
|
|
"step": 810
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12473565563559533,
|
||
|
|
"epoch": 1.312,
|
||
|
|
"grad_norm": 2.296875,
|
||
|
|
"learning_rate": 1.3122222222222222e-05,
|
||
|
|
"loss": 0.0738,
|
||
|
|
"mean_token_accuracy": 0.9808966059237718,
|
||
|
|
"num_tokens": 16028375.0,
|
||
|
|
"step": 820
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12206203136593104,
|
||
|
|
"epoch": 1.328,
|
||
|
|
"grad_norm": 2.265625,
|
||
|
|
"learning_rate": 1.3011111111111112e-05,
|
||
|
|
"loss": 0.0712,
|
||
|
|
"mean_token_accuracy": 0.9813739657402039,
|
||
|
|
"num_tokens": 16230555.0,
|
||
|
|
"step": 830
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11784348418004811,
|
||
|
|
"epoch": 1.3439999999999999,
|
||
|
|
"grad_norm": 2.359375,
|
||
|
|
"learning_rate": 1.2900000000000002e-05,
|
||
|
|
"loss": 0.0692,
|
||
|
|
"mean_token_accuracy": 0.9817070845514536,
|
||
|
|
"num_tokens": 16436297.0,
|
||
|
|
"step": 840
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12049585804343224,
|
||
|
|
"epoch": 1.3599999999999999,
|
||
|
|
"grad_norm": 2.234375,
|
||
|
|
"learning_rate": 1.278888888888889e-05,
|
||
|
|
"loss": 0.0719,
|
||
|
|
"mean_token_accuracy": 0.9809176307171583,
|
||
|
|
"num_tokens": 16630563.0,
|
||
|
|
"step": 850
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12437316868454218,
|
||
|
|
"epoch": 1.376,
|
||
|
|
"grad_norm": 2.8125,
|
||
|
|
"learning_rate": 1.2677777777777778e-05,
|
||
|
|
"loss": 0.0751,
|
||
|
|
"mean_token_accuracy": 0.9806294579058885,
|
||
|
|
"num_tokens": 16822602.0,
|
||
|
|
"step": 860
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12243083966895937,
|
||
|
|
"epoch": 1.392,
|
||
|
|
"grad_norm": 2.34375,
|
||
|
|
"learning_rate": 1.2566666666666668e-05,
|
||
|
|
"loss": 0.0729,
|
||
|
|
"mean_token_accuracy": 0.9806884177029133,
|
||
|
|
"num_tokens": 17015750.0,
|
||
|
|
"step": 870
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12194055491127073,
|
||
|
|
"epoch": 1.408,
|
||
|
|
"grad_norm": 2.21875,
|
||
|
|
"learning_rate": 1.2455555555555556e-05,
|
||
|
|
"loss": 0.0738,
|
||
|
|
"mean_token_accuracy": 0.9811285879462958,
|
||
|
|
"num_tokens": 17207248.0,
|
||
|
|
"step": 880
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12085925284773111,
|
||
|
|
"epoch": 1.424,
|
||
|
|
"grad_norm": 2.40625,
|
||
|
|
"learning_rate": 1.2344444444444446e-05,
|
||
|
|
"loss": 0.0714,
|
||
|
|
"mean_token_accuracy": 0.9814036220312119,
|
||
|
|
"num_tokens": 17403869.0,
|
||
|
|
"step": 890
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12371515287086368,
|
||
|
|
"epoch": 1.44,
|
||
|
|
"grad_norm": 2.671875,
|
||
|
|
"learning_rate": 1.2233333333333335e-05,
|
||
|
|
"loss": 0.0732,
|
||
|
|
"mean_token_accuracy": 0.9810383848845958,
|
||
|
|
"num_tokens": 17597428.0,
|
||
|
|
"step": 900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.44,
|
||
|
|
"eval_biology_entropy": 0.3943380331993103,
|
||
|
|
"eval_biology_loss": 2.181544542312622,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.6840425395965576,
|
||
|
|
"eval_biology_num_tokens": 17597428.0,
|
||
|
|
"eval_biology_runtime": 39.5119,
|
||
|
|
"eval_biology_samples_per_second": 12.654,
|
||
|
|
"eval_biology_steps_per_second": 3.164,
|
||
|
|
"step": 900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.44,
|
||
|
|
"eval_chemistry_entropy": 0.24567515325546266,
|
||
|
|
"eval_chemistry_loss": 1.3865065574645996,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7803217101097107,
|
||
|
|
"eval_chemistry_num_tokens": 17597428.0,
|
||
|
|
"eval_chemistry_runtime": 49.2083,
|
||
|
|
"eval_chemistry_samples_per_second": 10.161,
|
||
|
|
"eval_chemistry_steps_per_second": 2.54,
|
||
|
|
"step": 900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11841928996145726,
|
||
|
|
"epoch": 1.456,
|
||
|
|
"grad_norm": 2.421875,
|
||
|
|
"learning_rate": 1.2122222222222223e-05,
|
||
|
|
"loss": 0.0692,
|
||
|
|
"mean_token_accuracy": 0.9820894330739975,
|
||
|
|
"num_tokens": 17798083.0,
|
||
|
|
"step": 910
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12107395366765558,
|
||
|
|
"epoch": 1.472,
|
||
|
|
"grad_norm": 2.09375,
|
||
|
|
"learning_rate": 1.2011111111111111e-05,
|
||
|
|
"loss": 0.0727,
|
||
|
|
"mean_token_accuracy": 0.980775249004364,
|
||
|
|
"num_tokens": 17993802.0,
|
||
|
|
"step": 920
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11666952911764383,
|
||
|
|
"epoch": 1.488,
|
||
|
|
"grad_norm": 2.234375,
|
||
|
|
"learning_rate": 1.1900000000000001e-05,
|
||
|
|
"loss": 0.0711,
|
||
|
|
"mean_token_accuracy": 0.9814896911382676,
|
||
|
|
"num_tokens": 18185570.0,
|
||
|
|
"step": 930
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12296288679353892,
|
||
|
|
"epoch": 1.504,
|
||
|
|
"grad_norm": 2.203125,
|
||
|
|
"learning_rate": 1.178888888888889e-05,
|
||
|
|
"loss": 0.0721,
|
||
|
|
"mean_token_accuracy": 0.9813314471393824,
|
||
|
|
"num_tokens": 18377903.0,
|
||
|
|
"step": 940
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11539274472743273,
|
||
|
|
"epoch": 1.52,
|
||
|
|
"grad_norm": 2.21875,
|
||
|
|
"learning_rate": 1.1677777777777779e-05,
|
||
|
|
"loss": 0.0691,
|
||
|
|
"mean_token_accuracy": 0.9822911247611046,
|
||
|
|
"num_tokens": 18579284.0,
|
||
|
|
"step": 950
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12136667580343782,
|
||
|
|
"epoch": 1.536,
|
||
|
|
"grad_norm": 2.140625,
|
||
|
|
"learning_rate": 1.1566666666666669e-05,
|
||
|
|
"loss": 0.0725,
|
||
|
|
"mean_token_accuracy": 0.9812386609613896,
|
||
|
|
"num_tokens": 18765324.0,
|
||
|
|
"step": 960
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11671084221452474,
|
||
|
|
"epoch": 1.552,
|
||
|
|
"grad_norm": 2.15625,
|
||
|
|
"learning_rate": 1.1455555555555557e-05,
|
||
|
|
"loss": 0.0676,
|
||
|
|
"mean_token_accuracy": 0.9818577177822589,
|
||
|
|
"num_tokens": 18966382.0,
|
||
|
|
"step": 970
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11611056034453213,
|
||
|
|
"epoch": 1.568,
|
||
|
|
"grad_norm": 2.296875,
|
||
|
|
"learning_rate": 1.1344444444444445e-05,
|
||
|
|
"loss": 0.0701,
|
||
|
|
"mean_token_accuracy": 0.9818583976477384,
|
||
|
|
"num_tokens": 19160509.0,
|
||
|
|
"step": 980
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11777197350747884,
|
||
|
|
"epoch": 1.584,
|
||
|
|
"grad_norm": 2.1875,
|
||
|
|
"learning_rate": 1.1233333333333333e-05,
|
||
|
|
"loss": 0.0697,
|
||
|
|
"mean_token_accuracy": 0.9817087065428496,
|
||
|
|
"num_tokens": 19354662.0,
|
||
|
|
"step": 990
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.12017350159585476,
|
||
|
|
"epoch": 1.6,
|
||
|
|
"grad_norm": 2.28125,
|
||
|
|
"learning_rate": 1.1122222222222223e-05,
|
||
|
|
"loss": 0.0718,
|
||
|
|
"mean_token_accuracy": 0.9814203523099423,
|
||
|
|
"num_tokens": 19543088.0,
|
||
|
|
"step": 1000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.6,
|
||
|
|
"eval_biology_entropy": 0.3948675305843353,
|
||
|
|
"eval_biology_loss": 2.1792049407958984,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.6840984811782836,
|
||
|
|
"eval_biology_num_tokens": 19543088.0,
|
||
|
|
"eval_biology_runtime": 39.6119,
|
||
|
|
"eval_biology_samples_per_second": 12.622,
|
||
|
|
"eval_biology_steps_per_second": 3.156,
|
||
|
|
"step": 1000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.6,
|
||
|
|
"eval_chemistry_entropy": 0.24652056658267976,
|
||
|
|
"eval_chemistry_loss": 1.388296365737915,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7802323937416077,
|
||
|
|
"eval_chemistry_num_tokens": 19543088.0,
|
||
|
|
"eval_chemistry_runtime": 48.7687,
|
||
|
|
"eval_chemistry_samples_per_second": 10.252,
|
||
|
|
"eval_chemistry_steps_per_second": 2.563,
|
||
|
|
"step": 1000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11641141260042787,
|
||
|
|
"epoch": 1.616,
|
||
|
|
"grad_norm": 2.109375,
|
||
|
|
"learning_rate": 1.1011111111111113e-05,
|
||
|
|
"loss": 0.0685,
|
||
|
|
"mean_token_accuracy": 0.9818709082901478,
|
||
|
|
"num_tokens": 19743506.0,
|
||
|
|
"step": 1010
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11852226112969219,
|
||
|
|
"epoch": 1.6320000000000001,
|
||
|
|
"grad_norm": 2.40625,
|
||
|
|
"learning_rate": 1.0900000000000002e-05,
|
||
|
|
"loss": 0.0679,
|
||
|
|
"mean_token_accuracy": 0.9817365873605013,
|
||
|
|
"num_tokens": 19937567.0,
|
||
|
|
"step": 1020
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11748991990461946,
|
||
|
|
"epoch": 1.6480000000000001,
|
||
|
|
"grad_norm": 2.125,
|
||
|
|
"learning_rate": 1.0788888888888889e-05,
|
||
|
|
"loss": 0.0699,
|
||
|
|
"mean_token_accuracy": 0.9814462032169103,
|
||
|
|
"num_tokens": 20129743.0,
|
||
|
|
"step": 1030
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11318188132718206,
|
||
|
|
"epoch": 1.6640000000000001,
|
||
|
|
"grad_norm": 2.109375,
|
||
|
|
"learning_rate": 1.0677777777777779e-05,
|
||
|
|
"loss": 0.068,
|
||
|
|
"mean_token_accuracy": 0.9825476247817277,
|
||
|
|
"num_tokens": 20331548.0,
|
||
|
|
"step": 1040
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11528927914332598,
|
||
|
|
"epoch": 1.6800000000000002,
|
||
|
|
"grad_norm": 2.078125,
|
||
|
|
"learning_rate": 1.0566666666666667e-05,
|
||
|
|
"loss": 0.0684,
|
||
|
|
"mean_token_accuracy": 0.9821649711579085,
|
||
|
|
"num_tokens": 20524247.0,
|
||
|
|
"step": 1050
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11758545534685254,
|
||
|
|
"epoch": 1.696,
|
||
|
|
"grad_norm": 1.9296875,
|
||
|
|
"learning_rate": 1.0455555555555556e-05,
|
||
|
|
"loss": 0.0688,
|
||
|
|
"mean_token_accuracy": 0.9817702580243349,
|
||
|
|
"num_tokens": 20718121.0,
|
||
|
|
"step": 1060
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11215258438605816,
|
||
|
|
"epoch": 1.712,
|
||
|
|
"grad_norm": 2.109375,
|
||
|
|
"learning_rate": 1.0344444444444446e-05,
|
||
|
|
"loss": 0.066,
|
||
|
|
"mean_token_accuracy": 0.9826954051852226,
|
||
|
|
"num_tokens": 20921419.0,
|
||
|
|
"step": 1070
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11761863068677485,
|
||
|
|
"epoch": 1.728,
|
||
|
|
"grad_norm": 2.078125,
|
||
|
|
"learning_rate": 1.0233333333333336e-05,
|
||
|
|
"loss": 0.0673,
|
||
|
|
"mean_token_accuracy": 0.9817208234220743,
|
||
|
|
"num_tokens": 21118581.0,
|
||
|
|
"step": 1080
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11461647190153598,
|
||
|
|
"epoch": 1.744,
|
||
|
|
"grad_norm": 2.34375,
|
||
|
|
"learning_rate": 1.0122222222222222e-05,
|
||
|
|
"loss": 0.0683,
|
||
|
|
"mean_token_accuracy": 0.9821706589311361,
|
||
|
|
"num_tokens": 21315128.0,
|
||
|
|
"step": 1090
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11213490981608629,
|
||
|
|
"epoch": 1.76,
|
||
|
|
"grad_norm": 1.921875,
|
||
|
|
"learning_rate": 1.0011111111111112e-05,
|
||
|
|
"loss": 0.0668,
|
||
|
|
"mean_token_accuracy": 0.9826094429939985,
|
||
|
|
"num_tokens": 21511066.0,
|
||
|
|
"step": 1100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.76,
|
||
|
|
"eval_biology_entropy": 0.39157771730422974,
|
||
|
|
"eval_biology_loss": 2.183682680130005,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.6850721926689148,
|
||
|
|
"eval_biology_num_tokens": 21511066.0,
|
||
|
|
"eval_biology_runtime": 39.4964,
|
||
|
|
"eval_biology_samples_per_second": 12.659,
|
||
|
|
"eval_biology_steps_per_second": 3.165,
|
||
|
|
"step": 1100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.76,
|
||
|
|
"eval_chemistry_entropy": 0.24445027112960815,
|
||
|
|
"eval_chemistry_loss": 1.3950114250183105,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7797145857810974,
|
||
|
|
"eval_chemistry_num_tokens": 21511066.0,
|
||
|
|
"eval_chemistry_runtime": 49.2195,
|
||
|
|
"eval_chemistry_samples_per_second": 10.159,
|
||
|
|
"eval_chemistry_steps_per_second": 2.54,
|
||
|
|
"step": 1100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11297063264064491,
|
||
|
|
"epoch": 1.776,
|
||
|
|
"grad_norm": 2.078125,
|
||
|
|
"learning_rate": 9.9e-06,
|
||
|
|
"loss": 0.0678,
|
||
|
|
"mean_token_accuracy": 0.9824550054967404,
|
||
|
|
"num_tokens": 21704368.0,
|
||
|
|
"step": 1110
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11269787652418017,
|
||
|
|
"epoch": 1.792,
|
||
|
|
"grad_norm": 1.8359375,
|
||
|
|
"learning_rate": 9.78888888888889e-06,
|
||
|
|
"loss": 0.0652,
|
||
|
|
"mean_token_accuracy": 0.9826549135148526,
|
||
|
|
"num_tokens": 21905784.0,
|
||
|
|
"step": 1120
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11504443329758943,
|
||
|
|
"epoch": 1.808,
|
||
|
|
"grad_norm": 1.9765625,
|
||
|
|
"learning_rate": 9.677777777777778e-06,
|
||
|
|
"loss": 0.0673,
|
||
|
|
"mean_token_accuracy": 0.9823829177767038,
|
||
|
|
"num_tokens": 22094159.0,
|
||
|
|
"step": 1130
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11632196293212474,
|
||
|
|
"epoch": 1.8239999999999998,
|
||
|
|
"grad_norm": 1.9921875,
|
||
|
|
"learning_rate": 9.566666666666668e-06,
|
||
|
|
"loss": 0.0672,
|
||
|
|
"mean_token_accuracy": 0.9823333639651537,
|
||
|
|
"num_tokens": 22290654.0,
|
||
|
|
"step": 1140
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11296359319239854,
|
||
|
|
"epoch": 1.8399999999999999,
|
||
|
|
"grad_norm": 2.0625,
|
||
|
|
"learning_rate": 9.455555555555557e-06,
|
||
|
|
"loss": 0.0675,
|
||
|
|
"mean_token_accuracy": 0.9825551677495241,
|
||
|
|
"num_tokens": 22485675.0,
|
||
|
|
"step": 1150
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.10986362439580262,
|
||
|
|
"epoch": 1.8559999999999999,
|
||
|
|
"grad_norm": 2.265625,
|
||
|
|
"learning_rate": 9.344444444444446e-06,
|
||
|
|
"loss": 0.0637,
|
||
|
|
"mean_token_accuracy": 0.9831937052309513,
|
||
|
|
"num_tokens": 22689900.0,
|
||
|
|
"step": 1160
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11179361823014915,
|
||
|
|
"epoch": 1.8719999999999999,
|
||
|
|
"grad_norm": 1.828125,
|
||
|
|
"learning_rate": 9.233333333333334e-06,
|
||
|
|
"loss": 0.065,
|
||
|
|
"mean_token_accuracy": 0.9827314846217632,
|
||
|
|
"num_tokens": 22887275.0,
|
||
|
|
"step": 1170
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11337793134152889,
|
||
|
|
"epoch": 1.888,
|
||
|
|
"grad_norm": 1.875,
|
||
|
|
"learning_rate": 9.122222222222223e-06,
|
||
|
|
"loss": 0.0663,
|
||
|
|
"mean_token_accuracy": 0.9825703386217356,
|
||
|
|
"num_tokens": 23081241.0,
|
||
|
|
"step": 1180
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11284571103751659,
|
||
|
|
"epoch": 1.904,
|
||
|
|
"grad_norm": 2.296875,
|
||
|
|
"learning_rate": 9.011111111111111e-06,
|
||
|
|
"loss": 0.0655,
|
||
|
|
"mean_token_accuracy": 0.9827509298920631,
|
||
|
|
"num_tokens": 23276417.0,
|
||
|
|
"step": 1190
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.1114346832036972,
|
||
|
|
"epoch": 1.92,
|
||
|
|
"grad_norm": 1.9296875,
|
||
|
|
"learning_rate": 8.900000000000001e-06,
|
||
|
|
"loss": 0.065,
|
||
|
|
"mean_token_accuracy": 0.9832344222813845,
|
||
|
|
"num_tokens": 23476340.0,
|
||
|
|
"step": 1200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.92,
|
||
|
|
"eval_biology_entropy": 0.3916603057384491,
|
||
|
|
"eval_biology_loss": 2.1840121746063232,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.6849033751487732,
|
||
|
|
"eval_biology_num_tokens": 23476340.0,
|
||
|
|
"eval_biology_runtime": 39.5552,
|
||
|
|
"eval_biology_samples_per_second": 12.641,
|
||
|
|
"eval_biology_steps_per_second": 3.16,
|
||
|
|
"step": 1200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.92,
|
||
|
|
"eval_chemistry_entropy": 0.24395500665903092,
|
||
|
|
"eval_chemistry_loss": 1.3961467742919922,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7803040590286255,
|
||
|
|
"eval_chemistry_num_tokens": 23476340.0,
|
||
|
|
"eval_chemistry_runtime": 49.2524,
|
||
|
|
"eval_chemistry_samples_per_second": 10.152,
|
||
|
|
"eval_chemistry_steps_per_second": 2.538,
|
||
|
|
"step": 1200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.10977238831110299,
|
||
|
|
"epoch": 1.936,
|
||
|
|
"grad_norm": 2.078125,
|
||
|
|
"learning_rate": 8.788888888888891e-06,
|
||
|
|
"loss": 0.0642,
|
||
|
|
"mean_token_accuracy": 0.9830574735999107,
|
||
|
|
"num_tokens": 23673208.0,
|
||
|
|
"step": 1210
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.10960933691821992,
|
||
|
|
"epoch": 1.952,
|
||
|
|
"grad_norm": 1.7109375,
|
||
|
|
"learning_rate": 8.677777777777779e-06,
|
||
|
|
"loss": 0.0637,
|
||
|
|
"mean_token_accuracy": 0.9831824779510498,
|
||
|
|
"num_tokens": 23870993.0,
|
||
|
|
"step": 1220
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11105740028433501,
|
||
|
|
"epoch": 1.968,
|
||
|
|
"grad_norm": 1.8984375,
|
||
|
|
"learning_rate": 8.566666666666667e-06,
|
||
|
|
"loss": 0.0648,
|
||
|
|
"mean_token_accuracy": 0.9832170728594065,
|
||
|
|
"num_tokens": 24066234.0,
|
||
|
|
"step": 1230
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.10861094021238386,
|
||
|
|
"epoch": 1.984,
|
||
|
|
"grad_norm": 1.9296875,
|
||
|
|
"learning_rate": 8.455555555555555e-06,
|
||
|
|
"loss": 0.0642,
|
||
|
|
"mean_token_accuracy": 0.9830795601010323,
|
||
|
|
"num_tokens": 24262389.0,
|
||
|
|
"step": 1240
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.11186399965081364,
|
||
|
|
"epoch": 2.0,
|
||
|
|
"grad_norm": 2.15625,
|
||
|
|
"learning_rate": 8.344444444444445e-06,
|
||
|
|
"loss": 0.0659,
|
||
|
|
"mean_token_accuracy": 0.9825448121875524,
|
||
|
|
"num_tokens": 24455544.0,
|
||
|
|
"step": 1250
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.088846565480344,
|
||
|
|
"epoch": 2.016,
|
||
|
|
"grad_norm": 1.1328125,
|
||
|
|
"learning_rate": 8.233333333333335e-06,
|
||
|
|
"loss": 0.0505,
|
||
|
|
"mean_token_accuracy": 0.9856560029089451,
|
||
|
|
"num_tokens": 24645425.0,
|
||
|
|
"step": 1260
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08578005221206694,
|
||
|
|
"epoch": 2.032,
|
||
|
|
"grad_norm": 0.98828125,
|
||
|
|
"learning_rate": 8.122222222222223e-06,
|
||
|
|
"loss": 0.0495,
|
||
|
|
"mean_token_accuracy": 0.9862964119762182,
|
||
|
|
"num_tokens": 24835956.0,
|
||
|
|
"step": 1270
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08365524045657366,
|
||
|
|
"epoch": 2.048,
|
||
|
|
"grad_norm": 0.890625,
|
||
|
|
"learning_rate": 8.011111111111113e-06,
|
||
|
|
"loss": 0.0482,
|
||
|
|
"mean_token_accuracy": 0.9864422902464867,
|
||
|
|
"num_tokens": 25036470.0,
|
||
|
|
"step": 1280
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08349559165071696,
|
||
|
|
"epoch": 2.064,
|
||
|
|
"grad_norm": 0.8671875,
|
||
|
|
"learning_rate": 7.9e-06,
|
||
|
|
"loss": 0.0489,
|
||
|
|
"mean_token_accuracy": 0.9865494292229414,
|
||
|
|
"num_tokens": 25230959.0,
|
||
|
|
"step": 1290
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08540160392876714,
|
||
|
|
"epoch": 2.08,
|
||
|
|
"grad_norm": 0.98828125,
|
||
|
|
"learning_rate": 7.788888888888889e-06,
|
||
|
|
"loss": 0.0498,
|
||
|
|
"mean_token_accuracy": 0.9862173371016979,
|
||
|
|
"num_tokens": 25422625.0,
|
||
|
|
"step": 1300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.08,
|
||
|
|
"eval_biology_entropy": 0.37641260612010957,
|
||
|
|
"eval_biology_loss": 2.236539125442505,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.6860877084732055,
|
||
|
|
"eval_biology_num_tokens": 25422625.0,
|
||
|
|
"eval_biology_runtime": 39.5503,
|
||
|
|
"eval_biology_samples_per_second": 12.642,
|
||
|
|
"eval_biology_steps_per_second": 3.161,
|
||
|
|
"step": 1300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.08,
|
||
|
|
"eval_chemistry_entropy": 0.2347547177672386,
|
||
|
|
"eval_chemistry_loss": 1.4374333620071411,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7814928584098816,
|
||
|
|
"eval_chemistry_num_tokens": 25422625.0,
|
||
|
|
"eval_chemistry_runtime": 49.2536,
|
||
|
|
"eval_chemistry_samples_per_second": 10.152,
|
||
|
|
"eval_chemistry_steps_per_second": 2.538,
|
||
|
|
"step": 1300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08387249778024852,
|
||
|
|
"epoch": 2.096,
|
||
|
|
"grad_norm": 0.9296875,
|
||
|
|
"learning_rate": 7.677777777777778e-06,
|
||
|
|
"loss": 0.0496,
|
||
|
|
"mean_token_accuracy": 0.9860624361783266,
|
||
|
|
"num_tokens": 25614286.0,
|
||
|
|
"step": 1310
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08452012583147735,
|
||
|
|
"epoch": 2.112,
|
||
|
|
"grad_norm": 0.88671875,
|
||
|
|
"learning_rate": 7.566666666666667e-06,
|
||
|
|
"loss": 0.0489,
|
||
|
|
"mean_token_accuracy": 0.986203158274293,
|
||
|
|
"num_tokens": 25810384.0,
|
||
|
|
"step": 1320
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08262390322051942,
|
||
|
|
"epoch": 2.128,
|
||
|
|
"grad_norm": 0.84765625,
|
||
|
|
"learning_rate": 7.455555555555556e-06,
|
||
|
|
"loss": 0.0493,
|
||
|
|
"mean_token_accuracy": 0.9862308964133263,
|
||
|
|
"num_tokens": 26005047.0,
|
||
|
|
"step": 1330
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08097941202577204,
|
||
|
|
"epoch": 2.144,
|
||
|
|
"grad_norm": 1.1796875,
|
||
|
|
"learning_rate": 7.344444444444445e-06,
|
||
|
|
"loss": 0.0481,
|
||
|
|
"mean_token_accuracy": 0.9865331679582596,
|
||
|
|
"num_tokens": 26207169.0,
|
||
|
|
"step": 1340
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08371762945316732,
|
||
|
|
"epoch": 2.16,
|
||
|
|
"grad_norm": 0.90234375,
|
||
|
|
"learning_rate": 7.233333333333334e-06,
|
||
|
|
"loss": 0.0511,
|
||
|
|
"mean_token_accuracy": 0.985823930054903,
|
||
|
|
"num_tokens": 26402150.0,
|
||
|
|
"step": 1350
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08566265315748751,
|
||
|
|
"epoch": 2.176,
|
||
|
|
"grad_norm": 0.8125,
|
||
|
|
"learning_rate": 7.122222222222222e-06,
|
||
|
|
"loss": 0.0487,
|
||
|
|
"mean_token_accuracy": 0.9863390911370515,
|
||
|
|
"num_tokens": 26599464.0,
|
||
|
|
"step": 1360
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08372032032348216,
|
||
|
|
"epoch": 2.192,
|
||
|
|
"grad_norm": 0.9375,
|
||
|
|
"learning_rate": 7.011111111111112e-06,
|
||
|
|
"loss": 0.0503,
|
||
|
|
"mean_token_accuracy": 0.9858050804585219,
|
||
|
|
"num_tokens": 26790601.0,
|
||
|
|
"step": 1370
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08293257770128548,
|
||
|
|
"epoch": 2.208,
|
||
|
|
"grad_norm": 0.93359375,
|
||
|
|
"learning_rate": 6.9e-06,
|
||
|
|
"loss": 0.049,
|
||
|
|
"mean_token_accuracy": 0.9863026011735201,
|
||
|
|
"num_tokens": 26988598.0,
|
||
|
|
"step": 1380
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08349582962691784,
|
||
|
|
"epoch": 2.224,
|
||
|
|
"grad_norm": 0.76171875,
|
||
|
|
"learning_rate": 6.788888888888889e-06,
|
||
|
|
"loss": 0.0492,
|
||
|
|
"mean_token_accuracy": 0.9860329531133175,
|
||
|
|
"num_tokens": 27181933.0,
|
||
|
|
"step": 1390
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08361051741521805,
|
||
|
|
"epoch": 2.24,
|
||
|
|
"grad_norm": 0.89453125,
|
||
|
|
"learning_rate": 6.677777777777779e-06,
|
||
|
|
"loss": 0.0496,
|
||
|
|
"mean_token_accuracy": 0.985880882665515,
|
||
|
|
"num_tokens": 27378686.0,
|
||
|
|
"step": 1400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.24,
|
||
|
|
"eval_biology_entropy": 0.3741675862073898,
|
||
|
|
"eval_biology_loss": 2.242098569869995,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.6867531027793884,
|
||
|
|
"eval_biology_num_tokens": 27378686.0,
|
||
|
|
"eval_biology_runtime": 39.7857,
|
||
|
|
"eval_biology_samples_per_second": 12.567,
|
||
|
|
"eval_biology_steps_per_second": 3.142,
|
||
|
|
"step": 1400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.24,
|
||
|
|
"eval_chemistry_entropy": 0.23381740498542786,
|
||
|
|
"eval_chemistry_loss": 1.4444365501403809,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7810734486579896,
|
||
|
|
"eval_chemistry_num_tokens": 27378686.0,
|
||
|
|
"eval_chemistry_runtime": 49.2506,
|
||
|
|
"eval_chemistry_samples_per_second": 10.152,
|
||
|
|
"eval_chemistry_steps_per_second": 2.538,
|
||
|
|
"step": 1400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.0837663013022393,
|
||
|
|
"epoch": 2.2560000000000002,
|
||
|
|
"grad_norm": 0.91015625,
|
||
|
|
"learning_rate": 6.566666666666667e-06,
|
||
|
|
"loss": 0.0495,
|
||
|
|
"mean_token_accuracy": 0.9859327003359795,
|
||
|
|
"num_tokens": 27571908.0,
|
||
|
|
"step": 1410
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08184252071660011,
|
||
|
|
"epoch": 2.2720000000000002,
|
||
|
|
"grad_norm": 0.80859375,
|
||
|
|
"learning_rate": 6.455555555555556e-06,
|
||
|
|
"loss": 0.0485,
|
||
|
|
"mean_token_accuracy": 0.9862752124667168,
|
||
|
|
"num_tokens": 27770478.0,
|
||
|
|
"step": 1420
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.0840547483181581,
|
||
|
|
"epoch": 2.288,
|
||
|
|
"grad_norm": 0.9296875,
|
||
|
|
"learning_rate": 6.3444444444444454e-06,
|
||
|
|
"loss": 0.0495,
|
||
|
|
"mean_token_accuracy": 0.9857813119888306,
|
||
|
|
"num_tokens": 27965558.0,
|
||
|
|
"step": 1430
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08257126114331186,
|
||
|
|
"epoch": 2.304,
|
||
|
|
"grad_norm": 0.7734375,
|
||
|
|
"learning_rate": 6.2333333333333335e-06,
|
||
|
|
"loss": 0.0481,
|
||
|
|
"mean_token_accuracy": 0.9863984134048224,
|
||
|
|
"num_tokens": 28165624.0,
|
||
|
|
"step": 1440
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08313491414301097,
|
||
|
|
"epoch": 2.32,
|
||
|
|
"grad_norm": 0.9453125,
|
||
|
|
"learning_rate": 6.1222222222222224e-06,
|
||
|
|
"loss": 0.0492,
|
||
|
|
"mean_token_accuracy": 0.9862582210451365,
|
||
|
|
"num_tokens": 28363594.0,
|
||
|
|
"step": 1450
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08305449401959777,
|
||
|
|
"epoch": 2.336,
|
||
|
|
"grad_norm": 1.0234375,
|
||
|
|
"learning_rate": 6.011111111111112e-06,
|
||
|
|
"loss": 0.0495,
|
||
|
|
"mean_token_accuracy": 0.9860414322465658,
|
||
|
|
"num_tokens": 28558104.0,
|
||
|
|
"step": 1460
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08424469081219285,
|
||
|
|
"epoch": 2.352,
|
||
|
|
"grad_norm": 0.89453125,
|
||
|
|
"learning_rate": 5.9e-06,
|
||
|
|
"loss": 0.0509,
|
||
|
|
"mean_token_accuracy": 0.9856379702687263,
|
||
|
|
"num_tokens": 28747265.0,
|
||
|
|
"step": 1470
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08349237225484103,
|
||
|
|
"epoch": 2.368,
|
||
|
|
"grad_norm": 0.89453125,
|
||
|
|
"learning_rate": 5.788888888888889e-06,
|
||
|
|
"loss": 0.0488,
|
||
|
|
"mean_token_accuracy": 0.9860499382019043,
|
||
|
|
"num_tokens": 28944694.0,
|
||
|
|
"step": 1480
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08621921115554869,
|
||
|
|
"epoch": 2.384,
|
||
|
|
"grad_norm": 0.875,
|
||
|
|
"learning_rate": 5.677777777777779e-06,
|
||
|
|
"loss": 0.0506,
|
||
|
|
"mean_token_accuracy": 0.9857000894844532,
|
||
|
|
"num_tokens": 29136617.0,
|
||
|
|
"step": 1490
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08343604276888072,
|
||
|
|
"epoch": 2.4,
|
||
|
|
"grad_norm": 0.76953125,
|
||
|
|
"learning_rate": 5.566666666666667e-06,
|
||
|
|
"loss": 0.0493,
|
||
|
|
"mean_token_accuracy": 0.986125298216939,
|
||
|
|
"num_tokens": 29333526.0,
|
||
|
|
"step": 1500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.4,
|
||
|
|
"eval_biology_entropy": 0.37332249569892884,
|
||
|
|
"eval_biology_loss": 2.2458555698394775,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.6866299691200256,
|
||
|
|
"eval_biology_num_tokens": 29333526.0,
|
||
|
|
"eval_biology_runtime": 39.6765,
|
||
|
|
"eval_biology_samples_per_second": 12.602,
|
||
|
|
"eval_biology_steps_per_second": 3.15,
|
||
|
|
"step": 1500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.4,
|
||
|
|
"eval_chemistry_entropy": 0.2335134366750717,
|
||
|
|
"eval_chemistry_loss": 1.446710228919983,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7809953107833862,
|
||
|
|
"eval_chemistry_num_tokens": 29333526.0,
|
||
|
|
"eval_chemistry_runtime": 48.8671,
|
||
|
|
"eval_chemistry_samples_per_second": 10.232,
|
||
|
|
"eval_chemistry_steps_per_second": 2.558,
|
||
|
|
"step": 1500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08370916359126568,
|
||
|
|
"epoch": 2.416,
|
||
|
|
"grad_norm": 0.93359375,
|
||
|
|
"learning_rate": 5.455555555555556e-06,
|
||
|
|
"loss": 0.0487,
|
||
|
|
"mean_token_accuracy": 0.9860077023506164,
|
||
|
|
"num_tokens": 29534153.0,
|
||
|
|
"step": 1510
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08138901412021368,
|
||
|
|
"epoch": 2.432,
|
||
|
|
"grad_norm": 1.03125,
|
||
|
|
"learning_rate": 5.344444444444446e-06,
|
||
|
|
"loss": 0.0479,
|
||
|
|
"mean_token_accuracy": 0.9862620770931244,
|
||
|
|
"num_tokens": 29736571.0,
|
||
|
|
"step": 1520
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08320502748247235,
|
||
|
|
"epoch": 2.448,
|
||
|
|
"grad_norm": 1.1484375,
|
||
|
|
"learning_rate": 5.233333333333334e-06,
|
||
|
|
"loss": 0.0495,
|
||
|
|
"mean_token_accuracy": 0.9859858397394419,
|
||
|
|
"num_tokens": 29932423.0,
|
||
|
|
"step": 1530
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08031986611895263,
|
||
|
|
"epoch": 2.464,
|
||
|
|
"grad_norm": 1.203125,
|
||
|
|
"learning_rate": 5.122222222222223e-06,
|
||
|
|
"loss": 0.0479,
|
||
|
|
"mean_token_accuracy": 0.9863451778888702,
|
||
|
|
"num_tokens": 30137125.0,
|
||
|
|
"step": 1540
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08042037494014949,
|
||
|
|
"epoch": 2.48,
|
||
|
|
"grad_norm": 0.74609375,
|
||
|
|
"learning_rate": 5.011111111111111e-06,
|
||
|
|
"loss": 0.0484,
|
||
|
|
"mean_token_accuracy": 0.9864198382943868,
|
||
|
|
"num_tokens": 30341656.0,
|
||
|
|
"step": 1550
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08549262350425124,
|
||
|
|
"epoch": 2.496,
|
||
|
|
"grad_norm": 0.69140625,
|
||
|
|
"learning_rate": 4.9000000000000005e-06,
|
||
|
|
"loss": 0.0498,
|
||
|
|
"mean_token_accuracy": 0.985682675614953,
|
||
|
|
"num_tokens": 30533732.0,
|
||
|
|
"step": 1560
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08396681451704353,
|
||
|
|
"epoch": 2.512,
|
||
|
|
"grad_norm": 1.28125,
|
||
|
|
"learning_rate": 4.7888888888888894e-06,
|
||
|
|
"loss": 0.0494,
|
||
|
|
"mean_token_accuracy": 0.9859193779528141,
|
||
|
|
"num_tokens": 30730599.0,
|
||
|
|
"step": 1570
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08237075076904148,
|
||
|
|
"epoch": 2.528,
|
||
|
|
"grad_norm": 0.875,
|
||
|
|
"learning_rate": 4.677777777777778e-06,
|
||
|
|
"loss": 0.0487,
|
||
|
|
"mean_token_accuracy": 0.9857875861227512,
|
||
|
|
"num_tokens": 30927950.0,
|
||
|
|
"step": 1580
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08105518384836614,
|
||
|
|
"epoch": 2.544,
|
||
|
|
"grad_norm": 1.09375,
|
||
|
|
"learning_rate": 4.566666666666667e-06,
|
||
|
|
"loss": 0.0494,
|
||
|
|
"mean_token_accuracy": 0.9860106244683265,
|
||
|
|
"num_tokens": 31125667.0,
|
||
|
|
"step": 1590
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08552559302188456,
|
||
|
|
"epoch": 2.56,
|
||
|
|
"grad_norm": 0.8125,
|
||
|
|
"learning_rate": 4.455555555555555e-06,
|
||
|
|
"loss": 0.0508,
|
||
|
|
"mean_token_accuracy": 0.9855562083423137,
|
||
|
|
"num_tokens": 31315742.0,
|
||
|
|
"step": 1600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.56,
|
||
|
|
"eval_biology_entropy": 0.372919380903244,
|
||
|
|
"eval_biology_loss": 2.248323678970337,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.6864355983734131,
|
||
|
|
"eval_biology_num_tokens": 31315742.0,
|
||
|
|
"eval_biology_runtime": 39.5052,
|
||
|
|
"eval_biology_samples_per_second": 12.657,
|
||
|
|
"eval_biology_steps_per_second": 3.164,
|
||
|
|
"step": 1600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.56,
|
||
|
|
"eval_chemistry_entropy": 0.23327860993146896,
|
||
|
|
"eval_chemistry_loss": 1.4478672742843628,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7808715200424194,
|
||
|
|
"eval_chemistry_num_tokens": 31315742.0,
|
||
|
|
"eval_chemistry_runtime": 49.2432,
|
||
|
|
"eval_chemistry_samples_per_second": 10.154,
|
||
|
|
"eval_chemistry_steps_per_second": 2.538,
|
||
|
|
"step": 1600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08430088346358389,
|
||
|
|
"epoch": 2.576,
|
||
|
|
"grad_norm": 0.77734375,
|
||
|
|
"learning_rate": 4.344444444444445e-06,
|
||
|
|
"loss": 0.0507,
|
||
|
|
"mean_token_accuracy": 0.9855997852981091,
|
||
|
|
"num_tokens": 31505374.0,
|
||
|
|
"step": 1610
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08420886148232967,
|
||
|
|
"epoch": 2.592,
|
||
|
|
"grad_norm": 0.69921875,
|
||
|
|
"learning_rate": 4.233333333333334e-06,
|
||
|
|
"loss": 0.0494,
|
||
|
|
"mean_token_accuracy": 0.9857381645590066,
|
||
|
|
"num_tokens": 31701680.0,
|
||
|
|
"step": 1620
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08595787608064712,
|
||
|
|
"epoch": 2.608,
|
||
|
|
"grad_norm": 0.984375,
|
||
|
|
"learning_rate": 4.122222222222222e-06,
|
||
|
|
"loss": 0.0504,
|
||
|
|
"mean_token_accuracy": 0.9856818422675133,
|
||
|
|
"num_tokens": 31893221.0,
|
||
|
|
"step": 1630
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08458381698001176,
|
||
|
|
"epoch": 2.624,
|
||
|
|
"grad_norm": 0.921875,
|
||
|
|
"learning_rate": 4.011111111111111e-06,
|
||
|
|
"loss": 0.0503,
|
||
|
|
"mean_token_accuracy": 0.9855606179684401,
|
||
|
|
"num_tokens": 32087176.0,
|
||
|
|
"step": 1640
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08533325386233628,
|
||
|
|
"epoch": 2.64,
|
||
|
|
"grad_norm": 0.921875,
|
||
|
|
"learning_rate": 3.900000000000001e-06,
|
||
|
|
"loss": 0.05,
|
||
|
|
"mean_token_accuracy": 0.9852504417300224,
|
||
|
|
"num_tokens": 32280689.0,
|
||
|
|
"step": 1650
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.07994957917835563,
|
||
|
|
"epoch": 2.656,
|
||
|
|
"grad_norm": 0.81640625,
|
||
|
|
"learning_rate": 3.7888888888888893e-06,
|
||
|
|
"loss": 0.0478,
|
||
|
|
"mean_token_accuracy": 0.9864356804639101,
|
||
|
|
"num_tokens": 32483392.0,
|
||
|
|
"step": 1660
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08173574756365269,
|
||
|
|
"epoch": 2.672,
|
||
|
|
"grad_norm": 0.7578125,
|
||
|
|
"learning_rate": 3.6777777777777778e-06,
|
||
|
|
"loss": 0.048,
|
||
|
|
"mean_token_accuracy": 0.9862112771719694,
|
||
|
|
"num_tokens": 32683246.0,
|
||
|
|
"step": 1670
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.07958637131378055,
|
||
|
|
"epoch": 2.6879999999999997,
|
||
|
|
"grad_norm": 0.8046875,
|
||
|
|
"learning_rate": 3.566666666666667e-06,
|
||
|
|
"loss": 0.0474,
|
||
|
|
"mean_token_accuracy": 0.986869765073061,
|
||
|
|
"num_tokens": 32884993.0,
|
||
|
|
"step": 1680
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08010483384132386,
|
||
|
|
"epoch": 2.7039999999999997,
|
||
|
|
"grad_norm": 0.890625,
|
||
|
|
"learning_rate": 3.455555555555556e-06,
|
||
|
|
"loss": 0.0476,
|
||
|
|
"mean_token_accuracy": 0.9867513310164213,
|
||
|
|
"num_tokens": 33088388.0,
|
||
|
|
"step": 1690
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08223347943276167,
|
||
|
|
"epoch": 2.7199999999999998,
|
||
|
|
"grad_norm": 0.94140625,
|
||
|
|
"learning_rate": 3.3444444444444445e-06,
|
||
|
|
"loss": 0.049,
|
||
|
|
"mean_token_accuracy": 0.9861364755779505,
|
||
|
|
"num_tokens": 33287208.0,
|
||
|
|
"step": 1700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.7199999999999998,
|
||
|
|
"eval_biology_entropy": 0.37285495436191557,
|
||
|
|
"eval_biology_loss": 2.2501490116119385,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.6862905750274658,
|
||
|
|
"eval_biology_num_tokens": 33287208.0,
|
||
|
|
"eval_biology_runtime": 39.5673,
|
||
|
|
"eval_biology_samples_per_second": 12.637,
|
||
|
|
"eval_biology_steps_per_second": 3.159,
|
||
|
|
"step": 1700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.7199999999999998,
|
||
|
|
"eval_chemistry_entropy": 0.23335686594247818,
|
||
|
|
"eval_chemistry_loss": 1.4490605592727661,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7807522406578064,
|
||
|
|
"eval_chemistry_num_tokens": 33287208.0,
|
||
|
|
"eval_chemistry_runtime": 49.3763,
|
||
|
|
"eval_chemistry_samples_per_second": 10.126,
|
||
|
|
"eval_chemistry_steps_per_second": 2.532,
|
||
|
|
"step": 1700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.0847947500878945,
|
||
|
|
"epoch": 2.7359999999999998,
|
||
|
|
"grad_norm": 0.94921875,
|
||
|
|
"learning_rate": 3.2333333333333334e-06,
|
||
|
|
"loss": 0.0508,
|
||
|
|
"mean_token_accuracy": 0.9851035326719284,
|
||
|
|
"num_tokens": 33474750.0,
|
||
|
|
"step": 1710
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.0839041729690507,
|
||
|
|
"epoch": 2.752,
|
||
|
|
"grad_norm": 1.1796875,
|
||
|
|
"learning_rate": 3.1222222222222228e-06,
|
||
|
|
"loss": 0.0508,
|
||
|
|
"mean_token_accuracy": 0.9855955693870783,
|
||
|
|
"num_tokens": 33664302.0,
|
||
|
|
"step": 1720
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08604868296533823,
|
||
|
|
"epoch": 2.768,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 3.0111111111111113e-06,
|
||
|
|
"loss": 0.0505,
|
||
|
|
"mean_token_accuracy": 0.9854719124734401,
|
||
|
|
"num_tokens": 33854081.0,
|
||
|
|
"step": 1730
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08103844472207129,
|
||
|
|
"epoch": 2.784,
|
||
|
|
"grad_norm": 0.8046875,
|
||
|
|
"learning_rate": 2.9e-06,
|
||
|
|
"loss": 0.0484,
|
||
|
|
"mean_token_accuracy": 0.9861982863396406,
|
||
|
|
"num_tokens": 34053149.0,
|
||
|
|
"step": 1740
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08212419885676354,
|
||
|
|
"epoch": 2.8,
|
||
|
|
"grad_norm": 0.87109375,
|
||
|
|
"learning_rate": 2.788888888888889e-06,
|
||
|
|
"loss": 0.0479,
|
||
|
|
"mean_token_accuracy": 0.9861491050571203,
|
||
|
|
"num_tokens": 34254105.0,
|
||
|
|
"step": 1750
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08174366024322807,
|
||
|
|
"epoch": 2.816,
|
||
|
|
"grad_norm": 0.8359375,
|
||
|
|
"learning_rate": 2.677777777777778e-06,
|
||
|
|
"loss": 0.0483,
|
||
|
|
"mean_token_accuracy": 0.9865252107381821,
|
||
|
|
"num_tokens": 34453342.0,
|
||
|
|
"step": 1760
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08624515237752348,
|
||
|
|
"epoch": 2.832,
|
||
|
|
"grad_norm": 0.8984375,
|
||
|
|
"learning_rate": 2.566666666666667e-06,
|
||
|
|
"loss": 0.0508,
|
||
|
|
"mean_token_accuracy": 0.9852249193936586,
|
||
|
|
"num_tokens": 34644113.0,
|
||
|
|
"step": 1770
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08373226635158063,
|
||
|
|
"epoch": 2.848,
|
||
|
|
"grad_norm": 0.9765625,
|
||
|
|
"learning_rate": 2.455555555555556e-06,
|
||
|
|
"loss": 0.0509,
|
||
|
|
"mean_token_accuracy": 0.9856173813343048,
|
||
|
|
"num_tokens": 34835062.0,
|
||
|
|
"step": 1780
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.0821458526654169,
|
||
|
|
"epoch": 2.864,
|
||
|
|
"grad_norm": 0.671875,
|
||
|
|
"learning_rate": 2.3444444444444448e-06,
|
||
|
|
"loss": 0.049,
|
||
|
|
"mean_token_accuracy": 0.9862217791378498,
|
||
|
|
"num_tokens": 35032361.0,
|
||
|
|
"step": 1790
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.0835355838527903,
|
||
|
|
"epoch": 2.88,
|
||
|
|
"grad_norm": 0.984375,
|
||
|
|
"learning_rate": 2.2333333333333333e-06,
|
||
|
|
"loss": 0.0499,
|
||
|
|
"mean_token_accuracy": 0.9855714619159699,
|
||
|
|
"num_tokens": 35226653.0,
|
||
|
|
"step": 1800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.88,
|
||
|
|
"eval_biology_entropy": 0.37314171600341794,
|
||
|
|
"eval_biology_loss": 2.2503252029418945,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.6864723601341247,
|
||
|
|
"eval_biology_num_tokens": 35226653.0,
|
||
|
|
"eval_biology_runtime": 39.5746,
|
||
|
|
"eval_biology_samples_per_second": 12.634,
|
||
|
|
"eval_biology_steps_per_second": 3.159,
|
||
|
|
"step": 1800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.88,
|
||
|
|
"eval_chemistry_entropy": 0.23350779289007187,
|
||
|
|
"eval_chemistry_loss": 1.449589729309082,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7807379026412964,
|
||
|
|
"eval_chemistry_num_tokens": 35226653.0,
|
||
|
|
"eval_chemistry_runtime": 49.2909,
|
||
|
|
"eval_chemistry_samples_per_second": 10.144,
|
||
|
|
"eval_chemistry_steps_per_second": 2.536,
|
||
|
|
"step": 1800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08435964593663811,
|
||
|
|
"epoch": 2.896,
|
||
|
|
"grad_norm": 0.84375,
|
||
|
|
"learning_rate": 2.1222222222222226e-06,
|
||
|
|
"loss": 0.0501,
|
||
|
|
"mean_token_accuracy": 0.9855885654687881,
|
||
|
|
"num_tokens": 35421873.0,
|
||
|
|
"step": 1810
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08406312065199018,
|
||
|
|
"epoch": 2.912,
|
||
|
|
"grad_norm": 0.70703125,
|
||
|
|
"learning_rate": 2.011111111111111e-06,
|
||
|
|
"loss": 0.0497,
|
||
|
|
"mean_token_accuracy": 0.9859340194612741,
|
||
|
|
"num_tokens": 35614590.0,
|
||
|
|
"step": 1820
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08338865605182946,
|
||
|
|
"epoch": 2.928,
|
||
|
|
"grad_norm": 0.859375,
|
||
|
|
"learning_rate": 1.9000000000000002e-06,
|
||
|
|
"loss": 0.05,
|
||
|
|
"mean_token_accuracy": 0.9860526237636804,
|
||
|
|
"num_tokens": 35806494.0,
|
||
|
|
"step": 1830
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08239444589708,
|
||
|
|
"epoch": 2.944,
|
||
|
|
"grad_norm": 0.88671875,
|
||
|
|
"learning_rate": 1.788888888888889e-06,
|
||
|
|
"loss": 0.0494,
|
||
|
|
"mean_token_accuracy": 0.9860717881470918,
|
||
|
|
"num_tokens": 36001867.0,
|
||
|
|
"step": 1840
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08404006839264185,
|
||
|
|
"epoch": 2.96,
|
||
|
|
"grad_norm": 0.7265625,
|
||
|
|
"learning_rate": 1.6777777777777779e-06,
|
||
|
|
"loss": 0.0489,
|
||
|
|
"mean_token_accuracy": 0.9858945500105619,
|
||
|
|
"num_tokens": 36198806.0,
|
||
|
|
"step": 1850
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08261097352951766,
|
||
|
|
"epoch": 2.976,
|
||
|
|
"grad_norm": 1.0078125,
|
||
|
|
"learning_rate": 1.566666666666667e-06,
|
||
|
|
"loss": 0.0493,
|
||
|
|
"mean_token_accuracy": 0.9861804407089949,
|
||
|
|
"num_tokens": 36393098.0,
|
||
|
|
"step": 1860
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08398979662451893,
|
||
|
|
"epoch": 2.992,
|
||
|
|
"grad_norm": 1.15625,
|
||
|
|
"learning_rate": 1.4555555555555557e-06,
|
||
|
|
"loss": 0.0497,
|
||
|
|
"mean_token_accuracy": 0.9860749129205942,
|
||
|
|
"num_tokens": 36589074.0,
|
||
|
|
"step": 1870
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.0843541685724631,
|
||
|
|
"epoch": 3.008,
|
||
|
|
"grad_norm": 0.56640625,
|
||
|
|
"learning_rate": 1.3444444444444446e-06,
|
||
|
|
"loss": 0.0491,
|
||
|
|
"mean_token_accuracy": 0.985735047981143,
|
||
|
|
"num_tokens": 36778710.0,
|
||
|
|
"step": 1880
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08233831936959177,
|
||
|
|
"epoch": 3.024,
|
||
|
|
"grad_norm": 0.45703125,
|
||
|
|
"learning_rate": 1.2333333333333335e-06,
|
||
|
|
"loss": 0.0476,
|
||
|
|
"mean_token_accuracy": 0.985984667390585,
|
||
|
|
"num_tokens": 36966789.0,
|
||
|
|
"step": 1890
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08126567860599607,
|
||
|
|
"epoch": 3.04,
|
||
|
|
"grad_norm": 0.546875,
|
||
|
|
"learning_rate": 1.1222222222222222e-06,
|
||
|
|
"loss": 0.047,
|
||
|
|
"mean_token_accuracy": 0.9863237645477057,
|
||
|
|
"num_tokens": 37156529.0,
|
||
|
|
"step": 1900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.04,
|
||
|
|
"eval_biology_entropy": 0.37237140417099,
|
||
|
|
"eval_biology_loss": 2.251655101776123,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.68663911485672,
|
||
|
|
"eval_biology_num_tokens": 37156529.0,
|
||
|
|
"eval_biology_runtime": 39.9216,
|
||
|
|
"eval_biology_samples_per_second": 12.525,
|
||
|
|
"eval_biology_steps_per_second": 3.131,
|
||
|
|
"step": 1900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.04,
|
||
|
|
"eval_chemistry_entropy": 0.23310688173770905,
|
||
|
|
"eval_chemistry_loss": 1.450459361076355,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7806414008140564,
|
||
|
|
"eval_chemistry_num_tokens": 37156529.0,
|
||
|
|
"eval_chemistry_runtime": 49.2616,
|
||
|
|
"eval_chemistry_samples_per_second": 10.15,
|
||
|
|
"eval_chemistry_steps_per_second": 2.537,
|
||
|
|
"step": 1900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.07634997691493481,
|
||
|
|
"epoch": 3.056,
|
||
|
|
"grad_norm": 0.69140625,
|
||
|
|
"learning_rate": 1.0111111111111111e-06,
|
||
|
|
"loss": 0.046,
|
||
|
|
"mean_token_accuracy": 0.9868301976472139,
|
||
|
|
"num_tokens": 37351325.0,
|
||
|
|
"step": 1910
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.07988243501167744,
|
||
|
|
"epoch": 3.072,
|
||
|
|
"grad_norm": 0.55078125,
|
||
|
|
"learning_rate": 9.000000000000001e-07,
|
||
|
|
"loss": 0.047,
|
||
|
|
"mean_token_accuracy": 0.9863556236028671,
|
||
|
|
"num_tokens": 37541069.0,
|
||
|
|
"step": 1920
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.07848384822718799,
|
||
|
|
"epoch": 3.088,
|
||
|
|
"grad_norm": 0.55078125,
|
||
|
|
"learning_rate": 7.888888888888889e-07,
|
||
|
|
"loss": 0.0467,
|
||
|
|
"mean_token_accuracy": 0.9864400655031205,
|
||
|
|
"num_tokens": 37733440.0,
|
||
|
|
"step": 1930
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.07745544153731317,
|
||
|
|
"epoch": 3.104,
|
||
|
|
"grad_norm": 0.5703125,
|
||
|
|
"learning_rate": 6.777777777777779e-07,
|
||
|
|
"loss": 0.0445,
|
||
|
|
"mean_token_accuracy": 0.9870091959834099,
|
||
|
|
"num_tokens": 37933661.0,
|
||
|
|
"step": 1940
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.0795532870106399,
|
||
|
|
"epoch": 3.12,
|
||
|
|
"grad_norm": 0.56640625,
|
||
|
|
"learning_rate": 5.666666666666667e-07,
|
||
|
|
"loss": 0.0462,
|
||
|
|
"mean_token_accuracy": 0.9865683149546385,
|
||
|
|
"num_tokens": 38126905.0,
|
||
|
|
"step": 1950
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.08223398483823985,
|
||
|
|
"epoch": 3.136,
|
||
|
|
"grad_norm": 0.80078125,
|
||
|
|
"learning_rate": 4.5555555555555563e-07,
|
||
|
|
"loss": 0.047,
|
||
|
|
"mean_token_accuracy": 0.9862224843353034,
|
||
|
|
"num_tokens": 38316042.0,
|
||
|
|
"step": 1960
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.07699157926253974,
|
||
|
|
"epoch": 3.152,
|
||
|
|
"grad_norm": 0.59375,
|
||
|
|
"learning_rate": 3.444444444444445e-07,
|
||
|
|
"loss": 0.0453,
|
||
|
|
"mean_token_accuracy": 0.9869033012539148,
|
||
|
|
"num_tokens": 38514212.0,
|
||
|
|
"step": 1970
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.07890783820766956,
|
||
|
|
"epoch": 3.168,
|
||
|
|
"grad_norm": 0.5390625,
|
||
|
|
"learning_rate": 2.3333333333333336e-07,
|
||
|
|
"loss": 0.0459,
|
||
|
|
"mean_token_accuracy": 0.9864298563450575,
|
||
|
|
"num_tokens": 38708822.0,
|
||
|
|
"step": 1980
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.0802168081048876,
|
||
|
|
"epoch": 3.184,
|
||
|
|
"grad_norm": 0.58984375,
|
||
|
|
"learning_rate": 1.2222222222222225e-07,
|
||
|
|
"loss": 0.0468,
|
||
|
|
"mean_token_accuracy": 0.9863268151879311,
|
||
|
|
"num_tokens": 38900708.0,
|
||
|
|
"step": 1990
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.07599673259537668,
|
||
|
|
"epoch": 3.2,
|
||
|
|
"grad_norm": 0.478515625,
|
||
|
|
"learning_rate": 1.1111111111111112e-08,
|
||
|
|
"loss": 0.0446,
|
||
|
|
"mean_token_accuracy": 0.9873865962028503,
|
||
|
|
"num_tokens": 39102325.0,
|
||
|
|
"step": 2000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.2,
|
||
|
|
"eval_biology_entropy": 0.372367325425148,
|
||
|
|
"eval_biology_loss": 2.25302791595459,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.6864387440681458,
|
||
|
|
"eval_biology_num_tokens": 39102325.0,
|
||
|
|
"eval_biology_runtime": 39.5689,
|
||
|
|
"eval_biology_samples_per_second": 12.636,
|
||
|
|
"eval_biology_steps_per_second": 3.159,
|
||
|
|
"step": 2000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.2,
|
||
|
|
"eval_chemistry_entropy": 0.23305992305278778,
|
||
|
|
"eval_chemistry_loss": 1.4514989852905273,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7805765233039856,
|
||
|
|
"eval_chemistry_num_tokens": 39102325.0,
|
||
|
|
"eval_chemistry_runtime": 48.8599,
|
||
|
|
"eval_chemistry_samples_per_second": 10.233,
|
||
|
|
"eval_chemistry_steps_per_second": 2.558,
|
||
|
|
"step": 2000
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 10,
|
||
|
|
"max_steps": 2000,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 4,
|
||
|
|
"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": 3.279038413131436e+18,
|
||
|
|
"train_batch_size": 4,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|