Files
v53_public091_full_lr1e5_ep3/trainer_state.json
ModelHub XC b85b0fe15a 初始化项目,由ModelHub XC社区提供模型
Model: lldois/v53_public091_full_lr1e5_ep3
Source: Original Platform
2026-07-21 11:16:09 +08:00

723 lines
17 KiB
JSON

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"eval_steps": 500,
"global_step": 978,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03067484662576687,
"grad_norm": 27.375,
"learning_rate": 3e-06,
"loss": 2.8472824096679688,
"step": 10
},
{
"epoch": 0.06134969325153374,
"grad_norm": 10.0,
"learning_rate": 6.333333333333333e-06,
"loss": 2.606644058227539,
"step": 20
},
{
"epoch": 0.09202453987730061,
"grad_norm": 7.71875,
"learning_rate": 9.666666666666667e-06,
"loss": 2.3744272232055663,
"step": 30
},
{
"epoch": 0.12269938650306748,
"grad_norm": 3.859375,
"learning_rate": 9.997776301493914e-06,
"loss": 2.1879425048828125,
"step": 40
},
{
"epoch": 0.15337423312883436,
"grad_norm": 4.1875,
"learning_rate": 9.990091981794752e-06,
"loss": 1.9233184814453126,
"step": 50
},
{
"epoch": 0.18404907975460122,
"grad_norm": 3.46875,
"learning_rate": 9.976928024003827e-06,
"loss": 1.8812320709228516,
"step": 60
},
{
"epoch": 0.2147239263803681,
"grad_norm": 3.15625,
"learning_rate": 9.958298883510904e-06,
"loss": 1.849131965637207,
"step": 70
},
{
"epoch": 0.24539877300613497,
"grad_norm": 3.015625,
"learning_rate": 9.934225017042676e-06,
"loss": 1.830779457092285,
"step": 80
},
{
"epoch": 0.27607361963190186,
"grad_norm": 2.984375,
"learning_rate": 9.90473286019917e-06,
"loss": 1.7402576446533202,
"step": 90
},
{
"epoch": 0.3067484662576687,
"grad_norm": 3.953125,
"learning_rate": 9.869854798424709e-06,
"loss": 1.7090816497802734,
"step": 100
},
{
"epoch": 0.3374233128834356,
"grad_norm": 2.859375,
"learning_rate": 9.829629131445342e-06,
"loss": 1.750804328918457,
"step": 110
},
{
"epoch": 0.36809815950920244,
"grad_norm": 3.046875,
"learning_rate": 9.784100031211778e-06,
"loss": 1.6839963912963867,
"step": 120
},
{
"epoch": 0.3987730061349693,
"grad_norm": 2.828125,
"learning_rate": 9.733317493394004e-06,
"loss": 1.6797113418579102,
"step": 130
},
{
"epoch": 0.4294478527607362,
"grad_norm": 3.59375,
"learning_rate": 9.677337282480856e-06,
"loss": 1.65740909576416,
"step": 140
},
{
"epoch": 0.4601226993865031,
"grad_norm": 3.328125,
"learning_rate": 9.61622087054484e-06,
"loss": 1.7063955307006835,
"step": 150
},
{
"epoch": 0.49079754601226994,
"grad_norm": 3.21875,
"learning_rate": 9.550035369739416e-06,
"loss": 1.7848052978515625,
"step": 160
},
{
"epoch": 0.5214723926380368,
"grad_norm": 4.03125,
"learning_rate": 9.47885345860291e-06,
"loss": 1.681644821166992,
"step": 170
},
{
"epoch": 0.5521472392638037,
"grad_norm": 2.859375,
"learning_rate": 9.40275330224994e-06,
"loss": 1.6839426040649415,
"step": 180
},
{
"epoch": 0.5828220858895705,
"grad_norm": 4.5625,
"learning_rate": 9.32181846653802e-06,
"loss": 1.7123714447021485,
"step": 190
},
{
"epoch": 0.6134969325153374,
"grad_norm": 4.6875,
"learning_rate": 9.236137826303592e-06,
"loss": 1.5903639793395996,
"step": 200
},
{
"epoch": 0.6441717791411042,
"grad_norm": 2.78125,
"learning_rate": 9.145805467768251e-06,
"loss": 1.6954927444458008,
"step": 210
},
{
"epoch": 0.6748466257668712,
"grad_norm": 3.53125,
"learning_rate": 9.050920585222309e-06,
"loss": 1.6463920593261718,
"step": 220
},
{
"epoch": 0.7055214723926381,
"grad_norm": 3.03125,
"learning_rate": 8.951587372099199e-06,
"loss": 1.653399658203125,
"step": 230
},
{
"epoch": 0.7361963190184049,
"grad_norm": 3.109375,
"learning_rate": 8.847914906560287e-06,
"loss": 1.6198898315429688,
"step": 240
},
{
"epoch": 0.7668711656441718,
"grad_norm": 3.59375,
"learning_rate": 8.74001703171574e-06,
"loss": 1.5865962982177735,
"step": 250
},
{
"epoch": 0.7975460122699386,
"grad_norm": 3.09375,
"learning_rate": 8.628012230613001e-06,
"loss": 1.5989309310913087,
"step": 260
},
{
"epoch": 0.8282208588957055,
"grad_norm": 2.9375,
"learning_rate": 8.51202349613012e-06,
"loss": 1.6074316024780273,
"step": 270
},
{
"epoch": 0.8588957055214724,
"grad_norm": 2.859375,
"learning_rate": 8.392178195916832e-06,
"loss": 1.613131332397461,
"step": 280
},
{
"epoch": 0.8895705521472392,
"grad_norm": 3.171875,
"learning_rate": 8.268607932531682e-06,
"loss": 1.6734298706054687,
"step": 290
},
{
"epoch": 0.9202453987730062,
"grad_norm": 3.203125,
"learning_rate": 8.141448398928773e-06,
"loss": 1.553337574005127,
"step": 300
},
{
"epoch": 0.950920245398773,
"grad_norm": 2.546875,
"learning_rate": 8.010839229452843e-06,
"loss": 1.6621391296386718,
"step": 310
},
{
"epoch": 0.9815950920245399,
"grad_norm": 3.390625,
"learning_rate": 7.876923846506283e-06,
"loss": 1.585585117340088,
"step": 320
},
{
"epoch": 1.0122699386503067,
"grad_norm": 2.90625,
"learning_rate": 7.739849303056485e-06,
"loss": 1.5903179168701171,
"step": 330
},
{
"epoch": 1.0429447852760736,
"grad_norm": 2.359375,
"learning_rate": 7.599766121156436e-06,
"loss": 1.6181034088134765,
"step": 340
},
{
"epoch": 1.0736196319018405,
"grad_norm": 2.921875,
"learning_rate": 7.45682812665591e-06,
"loss": 1.6091724395751954,
"step": 350
},
{
"epoch": 1.1042944785276074,
"grad_norm": 3.1875,
"learning_rate": 7.3111922802847466e-06,
"loss": 1.5868614196777344,
"step": 360
},
{
"epoch": 1.1349693251533743,
"grad_norm": 3.421875,
"learning_rate": 7.163018505293703e-06,
"loss": 1.5639878273010255,
"step": 370
},
{
"epoch": 1.165644171779141,
"grad_norm": 3.109375,
"learning_rate": 7.012469511842149e-06,
"loss": 1.591905117034912,
"step": 380
},
{
"epoch": 1.196319018404908,
"grad_norm": 2.6875,
"learning_rate": 6.859710618325461e-06,
"loss": 1.5971322059631348,
"step": 390
},
{
"epoch": 1.2269938650306749,
"grad_norm": 3.15625,
"learning_rate": 6.704909569838281e-06,
"loss": 1.562277889251709,
"step": 400
},
{
"epoch": 1.2576687116564418,
"grad_norm": 2.375,
"learning_rate": 6.5482363539730286e-06,
"loss": 1.6139312744140626,
"step": 410
},
{
"epoch": 1.2883435582822087,
"grad_norm": 2.625,
"learning_rate": 6.389863014155906e-06,
"loss": 1.5223264694213867,
"step": 420
},
{
"epoch": 1.3190184049079754,
"grad_norm": 2.90625,
"learning_rate": 6.22996346072539e-06,
"loss": 1.5772048950195312,
"step": 430
},
{
"epoch": 1.3496932515337423,
"grad_norm": 2.5625,
"learning_rate": 6.068713279960659e-06,
"loss": 1.5874180793762207,
"step": 440
},
{
"epoch": 1.3803680981595092,
"grad_norm": 2.984375,
"learning_rate": 5.906289541269659e-06,
"loss": 1.5322503089904784,
"step": 450
},
{
"epoch": 1.4110429447852761,
"grad_norm": 2.625,
"learning_rate": 5.742870602748547e-06,
"loss": 1.5830337524414062,
"step": 460
},
{
"epoch": 1.441717791411043,
"grad_norm": 2.90625,
"learning_rate": 5.578635915326032e-06,
"loss": 1.6036342620849608,
"step": 470
},
{
"epoch": 1.4723926380368098,
"grad_norm": 2.484375,
"learning_rate": 5.413765825707633e-06,
"loss": 1.6447635650634767,
"step": 480
},
{
"epoch": 1.5030674846625767,
"grad_norm": 2.890625,
"learning_rate": 5.2484413783363335e-06,
"loss": 1.5753100395202637,
"step": 490
},
{
"epoch": 1.5337423312883436,
"grad_norm": 3.046875,
"learning_rate": 5.082844116587009e-06,
"loss": 1.5989506721496582,
"step": 500
},
{
"epoch": 1.5644171779141103,
"grad_norm": 3.25,
"learning_rate": 4.917155883412993e-06,
"loss": 1.5853590011596679,
"step": 510
},
{
"epoch": 1.5950920245398774,
"grad_norm": 2.75,
"learning_rate": 4.751558621663668e-06,
"loss": 1.609914779663086,
"step": 520
},
{
"epoch": 1.6257668711656441,
"grad_norm": 2.328125,
"learning_rate": 4.58623417429237e-06,
"loss": 1.5885044097900392,
"step": 530
},
{
"epoch": 1.656441717791411,
"grad_norm": 2.90625,
"learning_rate": 4.421364084673969e-06,
"loss": 1.567371940612793,
"step": 540
},
{
"epoch": 1.687116564417178,
"grad_norm": 2.953125,
"learning_rate": 4.257129397251453e-06,
"loss": 1.5381607055664062,
"step": 550
},
{
"epoch": 1.7177914110429446,
"grad_norm": 3.53125,
"learning_rate": 4.093710458730343e-06,
"loss": 1.545565128326416,
"step": 560
},
{
"epoch": 1.7484662576687118,
"grad_norm": 3.03125,
"learning_rate": 3.931286720039343e-06,
"loss": 1.5632963180541992,
"step": 570
},
{
"epoch": 1.7791411042944785,
"grad_norm": 3.0625,
"learning_rate": 3.7700365392746106e-06,
"loss": 1.5951272010803224,
"step": 580
},
{
"epoch": 1.8098159509202454,
"grad_norm": 3.265625,
"learning_rate": 3.610136985844096e-06,
"loss": 1.527869987487793,
"step": 590
},
{
"epoch": 1.8404907975460123,
"grad_norm": 2.765625,
"learning_rate": 3.451763646026971e-06,
"loss": 1.51753568649292,
"step": 600
},
{
"epoch": 1.871165644171779,
"grad_norm": 3.109375,
"learning_rate": 3.29509043016172e-06,
"loss": 1.5482216835021974,
"step": 610
},
{
"epoch": 1.9018404907975461,
"grad_norm": 2.609375,
"learning_rate": 3.1402893816745395e-06,
"loss": 1.5700338363647461,
"step": 620
},
{
"epoch": 1.9325153374233128,
"grad_norm": 3.3125,
"learning_rate": 2.9875304881578516e-06,
"loss": 1.5351737976074218,
"step": 630
},
{
"epoch": 1.9631901840490797,
"grad_norm": 2.28125,
"learning_rate": 2.8369814947062994e-06,
"loss": 1.6016332626342773,
"step": 640
},
{
"epoch": 1.9938650306748467,
"grad_norm": 2.609375,
"learning_rate": 2.688807719715254e-06,
"loss": 1.575319766998291,
"step": 650
},
{
"epoch": 2.0245398773006134,
"grad_norm": 3.515625,
"learning_rate": 2.543171873344092e-06,
"loss": 1.549346351623535,
"step": 660
},
{
"epoch": 2.0552147239263805,
"grad_norm": 3.46875,
"learning_rate": 2.4002338788435654e-06,
"loss": 1.582388210296631,
"step": 670
},
{
"epoch": 2.085889570552147,
"grad_norm": 2.984375,
"learning_rate": 2.260150696943516e-06,
"loss": 1.542292881011963,
"step": 680
},
{
"epoch": 2.116564417177914,
"grad_norm": 3.703125,
"learning_rate": 2.123076153493717e-06,
"loss": 1.5415298461914062,
"step": 690
},
{
"epoch": 2.147239263803681,
"grad_norm": 2.921875,
"learning_rate": 1.989160770547159e-06,
"loss": 1.5674308776855468,
"step": 700
},
{
"epoch": 2.1779141104294477,
"grad_norm": 2.890625,
"learning_rate": 1.8585516010712284e-06,
"loss": 1.5271636962890625,
"step": 710
},
{
"epoch": 2.208588957055215,
"grad_norm": 2.578125,
"learning_rate": 1.7313920674683181e-06,
"loss": 1.5421697616577148,
"step": 720
},
{
"epoch": 2.2392638036809815,
"grad_norm": 3.171875,
"learning_rate": 1.6078218040831678e-06,
"loss": 1.5656076431274415,
"step": 730
},
{
"epoch": 2.2699386503067487,
"grad_norm": 3.359375,
"learning_rate": 1.4879765038698819e-06,
"loss": 1.4618969917297364,
"step": 740
},
{
"epoch": 2.3006134969325154,
"grad_norm": 2.71875,
"learning_rate": 1.371987769387001e-06,
"loss": 1.595311450958252,
"step": 750
},
{
"epoch": 2.331288343558282,
"grad_norm": 2.59375,
"learning_rate": 1.2599829682842618e-06,
"loss": 1.6427331924438477,
"step": 760
},
{
"epoch": 2.361963190184049,
"grad_norm": 2.765625,
"learning_rate": 1.1520850934397143e-06,
"loss": 1.5959955215454102,
"step": 770
},
{
"epoch": 2.392638036809816,
"grad_norm": 2.875,
"learning_rate": 1.0484126279008028e-06,
"loss": 1.5941368103027345,
"step": 780
},
{
"epoch": 2.4233128834355826,
"grad_norm": 2.734375,
"learning_rate": 9.490794147776927e-07,
"loss": 1.556960678100586,
"step": 790
},
{
"epoch": 2.4539877300613497,
"grad_norm": 3.046875,
"learning_rate": 8.541945322317507e-07,
"loss": 1.6013805389404296,
"step": 800
},
{
"epoch": 2.4846625766871164,
"grad_norm": 2.484375,
"learning_rate": 7.638621736964075e-07,
"loss": 1.530329704284668,
"step": 810
},
{
"epoch": 2.5153374233128836,
"grad_norm": 2.953125,
"learning_rate": 6.781815334619812e-07,
"loss": 1.5487473487854004,
"step": 820
},
{
"epoch": 2.5460122699386503,
"grad_norm": 3.296875,
"learning_rate": 5.972466977500613e-07,
"loss": 1.5803820610046386,
"step": 830
},
{
"epoch": 2.5766871165644174,
"grad_norm": 3.328125,
"learning_rate": 5.211465413970901e-07,
"loss": 1.59120512008667,
"step": 840
},
{
"epoch": 2.607361963190184,
"grad_norm": 2.921875,
"learning_rate": 4.4996463026058476e-07,
"loss": 1.586211585998535,
"step": 850
},
{
"epoch": 2.638036809815951,
"grad_norm": 2.9375,
"learning_rate": 3.837791294551624e-07,
"loss": 1.5542678833007812,
"step": 860
},
{
"epoch": 2.668711656441718,
"grad_norm": 2.53125,
"learning_rate": 3.2266271751914603e-07,
"loss": 1.5309386253356934,
"step": 870
},
{
"epoch": 2.6993865030674846,
"grad_norm": 3.140625,
"learning_rate": 2.666825066059986e-07,
"loss": 1.53997859954834,
"step": 880
},
{
"epoch": 2.7300613496932513,
"grad_norm": 2.203125,
"learning_rate": 2.1589996878822296e-07,
"loss": 1.590695858001709,
"step": 890
},
{
"epoch": 2.7607361963190185,
"grad_norm": 3.359375,
"learning_rate": 1.7037086855465902e-07,
"loss": 1.5309554100036622,
"step": 900
},
{
"epoch": 2.791411042944785,
"grad_norm": 3.203125,
"learning_rate": 1.3014520157529244e-07,
"loss": 1.5762608528137207,
"step": 910
},
{
"epoch": 2.8220858895705523,
"grad_norm": 3.59375,
"learning_rate": 9.526713980083092e-08,
"loss": 1.5450604438781739,
"step": 920
},
{
"epoch": 2.852760736196319,
"grad_norm": 2.859375,
"learning_rate": 6.577498295732454e-08,
"loss": 1.5562796592712402,
"step": 930
},
{
"epoch": 2.883435582822086,
"grad_norm": 3.0625,
"learning_rate": 4.170111648909736e-08,
"loss": 1.527918243408203,
"step": 940
},
{
"epoch": 2.914110429447853,
"grad_norm": 2.734375,
"learning_rate": 2.3071975996173392e-08,
"loss": 1.5454301834106445,
"step": 950
},
{
"epoch": 2.9447852760736195,
"grad_norm": 2.890625,
"learning_rate": 9.90801820524967e-09,
"loss": 1.5550783157348633,
"step": 960
},
{
"epoch": 2.9754601226993866,
"grad_norm": 3.328125,
"learning_rate": 2.223698506088612e-09,
"loss": 1.5878689765930176,
"step": 970
},
{
"epoch": 3.0,
"step": 978,
"total_flos": 4.689887540417925e+17,
"train_loss": 1.6425541134699722,
"train_runtime": 7896.9762,
"train_samples_per_second": 0.495,
"train_steps_per_second": 0.124
}
],
"logging_steps": 10,
"max_steps": 978,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": false,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 4.689887540417925e+17,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}