roonbug/11sivxlz
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

初始化项目，由ModelHub XC社区提供模型

Browse Source 
Model: roonbug/11sivxlz
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-09-07 15:58:16 +08:00
commit 8398654810
82 changed files with 19660 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

644
checkpoint-500/trainer_state.json Normal file
View File

@@ -0,0 +1,644 @@
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.8,
"eval_steps": 100,
"global_step": 500,
"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
}
],
"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": false
},
"attributes": {}
}
},
"total_flos": 8.165294965933916e+17,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}