Files
AfriqueQwen-14B-multiturn/trainer_state.json
ModelHub XC 031ff14420 初始化项目,由ModelHub XC社区提供模型
Model: israel/AfriqueQwen-14B-multiturn
Source: Original Platform
2026-04-29 12:28:53 +08:00

282 lines
7.2 KiB
JSON

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 5.0,
"eval_steps": 500,
"global_step": 34265,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.1459321415541773,
"grad_norm": 2.2002170436333888,
"learning_rate": 2.915086081120514e-06,
"loss": 1.091951904296875,
"step": 1000
},
{
"epoch": 0.2918642831083546,
"grad_norm": 2.3899237660554205,
"learning_rate": 5.833090166326233e-06,
"loss": 0.9010311279296875,
"step": 2000
},
{
"epoch": 0.43779642466253194,
"grad_norm": 1.7207455491966865,
"learning_rate": 8.751094251531953e-06,
"loss": 0.8496372680664063,
"step": 3000
},
{
"epoch": 0.5837285662167092,
"grad_norm": 1.49997561253202,
"learning_rate": 9.991513345767592e-06,
"loss": 0.8293939208984376,
"step": 4000
},
{
"epoch": 0.7296607077708865,
"grad_norm": 1.43919873990257,
"learning_rate": 9.936020028278053e-06,
"loss": 0.8073157348632812,
"step": 5000
},
{
"epoch": 0.8755928493250639,
"grad_norm": 1.384519444393501,
"learning_rate": 9.829343371836088e-06,
"loss": 0.7829708251953125,
"step": 6000
},
{
"epoch": 1.021452024808464,
"grad_norm": 1.457416645249146,
"learning_rate": 9.672589544454328e-06,
"loss": 0.7496401977539062,
"step": 7000
},
{
"epoch": 1.1673841663626414,
"grad_norm": 1.5133027310610572,
"learning_rate": 9.46738398205746e-06,
"loss": 0.6121725463867187,
"step": 8000
},
{
"epoch": 1.3133163079168186,
"grad_norm": 1.3855612557611925,
"learning_rate": 9.215854533761766e-06,
"loss": 0.611597412109375,
"step": 9000
},
{
"epoch": 1.459248449470996,
"grad_norm": 1.5802706606106216,
"learning_rate": 8.920609397454381e-06,
"loss": 0.6123408203125,
"step": 10000
},
{
"epoch": 1.6051805910251733,
"grad_norm": 1.4685873474216664,
"learning_rate": 8.584710074466158e-06,
"loss": 0.609525390625,
"step": 11000
},
{
"epoch": 1.7511127325793505,
"grad_norm": 1.219859265372452,
"learning_rate": 8.211639623780629e-06,
"loss": 0.6133668823242188,
"step": 12000
},
{
"epoch": 1.897044874133528,
"grad_norm": 1.2535196330309815,
"learning_rate": 7.805266544962458e-06,
"loss": 0.615207763671875,
"step": 13000
},
{
"epoch": 2.042904049616928,
"grad_norm": 1.5681754211680612,
"learning_rate": 7.3698046643160645e-06,
"loss": 0.5413321533203125,
"step": 14000
},
{
"epoch": 2.1888361911711054,
"grad_norm": 1.2447962924993787,
"learning_rate": 6.909769440229038e-06,
"loss": 0.38200238037109374,
"step": 15000
},
{
"epoch": 2.334768332725283,
"grad_norm": 1.4067667074111327,
"learning_rate": 6.4299311407857035e-06,
"loss": 0.38497344970703123,
"step": 16000
},
{
"epoch": 2.48070047427946,
"grad_norm": 1.326059120500027,
"learning_rate": 5.935265379168761e-06,
"loss": 0.3850032653808594,
"step": 17000
},
{
"epoch": 2.6266326158336373,
"grad_norm": 1.6923974019115828,
"learning_rate": 5.430901519764892e-06,
"loss": 0.38166015625,
"step": 18000
},
{
"epoch": 2.7725647573878147,
"grad_norm": 1.5367782861850559,
"learning_rate": 4.9220694899697185e-06,
"loss": 0.3862608642578125,
"step": 19000
},
{
"epoch": 2.918496898941992,
"grad_norm": 1.3054321737811847,
"learning_rate": 4.414045549219315e-06,
"loss": 0.38137054443359375,
"step": 20000
},
{
"epoch": 3.064356074425392,
"grad_norm": 1.69275953279231,
"learning_rate": 3.912097577588397e-06,
"loss": 0.2938824462890625,
"step": 21000
},
{
"epoch": 3.2102882159795696,
"grad_norm": 1.5240170572509457,
"learning_rate": 3.4214304512770823e-06,
"loss": 0.19125808715820314,
"step": 22000
},
{
"epoch": 3.356220357533747,
"grad_norm": 1.5774125509908914,
"learning_rate": 2.9471320714071095e-06,
"loss": 0.18847378540039061,
"step": 23000
},
{
"epoch": 3.502152499087924,
"grad_norm": 1.8421166872943977,
"learning_rate": 2.4941206057740675e-06,
"loss": 0.19000143432617186,
"step": 24000
},
{
"epoch": 3.6480846406421015,
"grad_norm": 1.5694212047061644,
"learning_rate": 2.06709349062457e-06,
"loss": 0.18782159423828124,
"step": 25000
},
{
"epoch": 3.7940167821962785,
"grad_norm": 1.496800606594422,
"learning_rate": 1.6704787212769829e-06,
"loss": 0.18233416748046874,
"step": 26000
},
{
"epoch": 3.939948923750456,
"grad_norm": 1.984744582152809,
"learning_rate": 1.3083889366705216e-06,
"loss": 0.17911607360839843,
"step": 27000
},
{
"epoch": 4.085808099233856,
"grad_norm": 1.2567953809414274,
"learning_rate": 9.845787739562829e-07,
"loss": 0.12317549133300781,
"step": 28000
},
{
"epoch": 4.231740240788033,
"grad_norm": 1.091077915720754,
"learning_rate": 7.024059353355333e-07,
"loss": 0.08010615539550782,
"step": 29000
},
{
"epoch": 4.377672382342211,
"grad_norm": 1.2460581955478622,
"learning_rate": 4.64796370857008e-07,
"loss": 0.08023818969726562,
"step": 30000
},
{
"epoch": 4.523604523896388,
"grad_norm": 1.7839106451867877,
"learning_rate": 2.7421393820510846e-07,
"loss": 0.07947054290771484,
"step": 31000
},
{
"epoch": 4.669536665450566,
"grad_norm": 1.1265268481087043,
"learning_rate": 1.326348540874095e-07,
"loss": 0.07735189056396484,
"step": 32000
},
{
"epoch": 4.815468807004743,
"grad_norm": 1.0896372619502366,
"learning_rate": 4.152720214406214e-08,
"loss": 0.07909833526611328,
"step": 33000
},
{
"epoch": 4.96140094855892,
"grad_norm": 1.2750658260191035,
"learning_rate": 1.8357098688476238e-09,
"loss": 0.07906644439697266,
"step": 34000
},
{
"epoch": 5.0,
"step": 34265,
"total_flos": 1091733654863872.0,
"train_loss": 0.42485430567312915,
"train_runtime": 114160.2816,
"train_samples_per_second": 2.401,
"train_steps_per_second": 0.3
}
],
"logging_steps": 1000,
"max_steps": 34265,
"num_input_tokens_seen": 0,
"num_train_epochs": 5,
"save_steps": 50000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 1091733654863872.0,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}