shubhamrgandhi/qwen3-8b-full-sft-prm-opus-distill-32k-lr5e6-flattened
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
qwen3-8b-full-sft-prm-opus-…/trainer_state.json
ModelHub XC ceba279c82 初始化项目，由ModelHub XC社区提供模型 
Model: shubhamrgandhi/qwen3-8b-full-sft-prm-opus-distill-32k-lr5e6-flattened
Source: Original Platform
2026-06-18 18:29:20 +08:00

506 lines
12 KiB
JSON
Raw Permalink Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"eval_steps": 500,
"global_step": 660,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.045454545454545456,
"grad_norm": 12.912503242492676,
"learning_rate": 6.818181818181818e-07,
"loss": 1.0602,
"step": 10
},
{
"epoch": 0.09090909090909091,
"grad_norm": 4.66077995300293,
"learning_rate": 1.4393939393939396e-06,
"loss": 0.9686,
"step": 20
},
{
"epoch": 0.13636363636363635,
"grad_norm": 2.7883694171905518,
"learning_rate": 2.196969696969697e-06,
"loss": 0.8175,
"step": 30
},
{
"epoch": 0.18181818181818182,
"grad_norm": 2.6756789684295654,
"learning_rate": 2.954545454545455e-06,
"loss": 0.7458,
"step": 40
},
{
"epoch": 0.22727272727272727,
"grad_norm": 2.839015245437622,
"learning_rate": 3.7121212121212124e-06,
"loss": 0.6929,
"step": 50
},
{
"epoch": 0.2727272727272727,
"grad_norm": 2.581166982650757,
"learning_rate": 4.46969696969697e-06,
"loss": 0.6465,
"step": 60
},
{
"epoch": 0.3181818181818182,
"grad_norm": 2.901815414428711,
"learning_rate": 4.999685319184688e-06,
"loss": 0.5931,
"step": 70
},
{
"epoch": 0.36363636363636365,
"grad_norm": 2.5252325534820557,
"learning_rate": 4.994093197099587e-06,
"loss": 0.6159,
"step": 80
},
{
"epoch": 0.4090909090909091,
"grad_norm": 2.8176751136779785,
"learning_rate": 4.981526170026621e-06,
"loss": 0.6072,
"step": 90
},
{
"epoch": 0.45454545454545453,
"grad_norm": 2.517207622528076,
"learning_rate": 4.962019382530521e-06,
"loss": 0.6049,
"step": 100
},
{
"epoch": 0.5,
"grad_norm": 2.6463892459869385,
"learning_rate": 4.935627386698418e-06,
"loss": 0.5677,
"step": 110
},
{
"epoch": 0.5454545454545454,
"grad_norm": 3.6619322299957275,
"learning_rate": 4.902423989581143e-06,
"loss": 0.5668,
"step": 120
},
{
"epoch": 0.5909090909090909,
"grad_norm": 2.821599245071411,
"learning_rate": 4.862502046786671e-06,
"loss": 0.5495,
"step": 130
},
{
"epoch": 0.6363636363636364,
"grad_norm": 2.5070018768310547,
"learning_rate": 4.815973202802966e-06,
"loss": 0.5233,
"step": 140
},
{
"epoch": 0.6818181818181818,
"grad_norm": 2.5705134868621826,
"learning_rate": 4.762967578776406e-06,
"loss": 0.5896,
"step": 150
},
{
"epoch": 0.7272727272727273,
"grad_norm": 2.4343154430389404,
"learning_rate": 4.703633408618955e-06,
"loss": 0.5506,
"step": 160
},
{
"epoch": 0.7727272727272727,
"grad_norm": 2.4548699855804443,
"learning_rate": 4.638136624461723e-06,
"loss": 0.5161,
"step": 170
},
{
"epoch": 0.8181818181818182,
"grad_norm": 2.902329683303833,
"learning_rate": 4.566660392614229e-06,
"loss": 0.569,
"step": 180
},
{
"epoch": 0.8636363636363636,
"grad_norm": 2.5096774101257324,
"learning_rate": 4.489404601327081e-06,
"loss": 0.555,
"step": 190
},
{
"epoch": 0.9090909090909091,
"grad_norm": 2.590094566345215,
"learning_rate": 4.406585301790595e-06,
"loss": 0.5055,
"step": 200
},
{
"epoch": 0.9545454545454546,
"grad_norm": 2.6825978755950928,
"learning_rate": 4.318434103932622e-06,
"loss": 0.5378,
"step": 210
},
{
"epoch": 1.0,
"grad_norm": 3.1130728721618652,
"learning_rate": 4.2251975287052804e-06,
"loss": 0.4894,
"step": 220
},
{
"epoch": 1.0454545454545454,
"grad_norm": 2.508296012878418,
"learning_rate": 4.127136318671984e-06,
"loss": 0.3795,
"step": 230
},
{
"epoch": 1.0909090909090908,
"grad_norm": 2.275791883468628,
"learning_rate": 4.024524708822739e-06,
"loss": 0.3842,
"step": 240
},
{
"epoch": 1.1363636363636362,
"grad_norm": 2.366997718811035,
"learning_rate": 3.917649659656927e-06,
"loss": 0.3773,
"step": 250
},
{
"epoch": 1.1818181818181819,
"grad_norm": 2.7410788536071777,
"learning_rate": 3.8068100546783315e-06,
"loss": 0.3814,
"step": 260
},
{
"epoch": 1.2272727272727273,
"grad_norm": 2.7614405155181885,
"learning_rate": 3.692315864546635e-06,
"loss": 0.3861,
"step": 270
},
{
"epoch": 1.2727272727272727,
"grad_norm": 2.2853896617889404,
"learning_rate": 3.5744872802229296e-06,
"loss": 0.401,
"step": 280
},
{
"epoch": 1.3181818181818181,
"grad_norm": 2.896815299987793,
"learning_rate": 3.453653817533435e-06,
"loss": 0.3942,
"step": 290
},
{
"epoch": 1.3636363636363638,
"grad_norm": 2.7985401153564453,
"learning_rate": 3.3301533956555886e-06,
"loss": 0.3784,
"step": 300
},
{
"epoch": 1.4090909090909092,
"grad_norm": 3.265619993209839,
"learning_rate": 3.2043313921035747e-06,
"loss": 0.4004,
"step": 310
},
{
"epoch": 1.4545454545454546,
"grad_norm": 2.6088955402374268,
"learning_rate": 3.0765396768561005e-06,
"loss": 0.3663,
"step": 320
},
{
"epoch": 1.5,
"grad_norm": 2.830472707748413,
"learning_rate": 2.9471356283275444e-06,
"loss": 0.4158,
"step": 330
},
{
"epoch": 1.5454545454545454,
"grad_norm": 2.780045509338379,
"learning_rate": 2.8164811339343736e-06,
"loss": 0.3799,
"step": 340
},
{
"epoch": 1.5909090909090908,
"grad_norm": 2.4751908779144287,
"learning_rate": 2.684941578051836e-06,
"loss": 0.3974,
"step": 350
},
{
"epoch": 1.6363636363636362,
"grad_norm": 2.8256654739379883,
"learning_rate": 2.5528848201911543e-06,
"loss": 0.3846,
"step": 360
},
{
"epoch": 1.6818181818181817,
"grad_norm": 2.9007036685943604,
"learning_rate": 2.4206801662548314e-06,
"loss": 0.3881,
"step": 370
},
{
"epoch": 1.7272727272727273,
"grad_norm": 2.53434157371521,
"learning_rate": 2.288697335747027e-06,
"loss": 0.3729,
"step": 380
},
{
"epoch": 1.7727272727272727,
"grad_norm": 3.7612838745117188,
"learning_rate": 2.157305427827264e-06,
"loss": 0.3735,
"step": 390
},
{
"epoch": 1.8181818181818183,
"grad_norm": 2.661956310272217,
"learning_rate": 2.0268718890989754e-06,
"loss": 0.3767,
"step": 400
},
{
"epoch": 1.8636363636363638,
"grad_norm": 2.765303611755371,
"learning_rate": 1.8977614860195297e-06,
"loss": 0.3745,
"step": 410
},
{
"epoch": 1.9090909090909092,
"grad_norm": 2.670576810836792,
"learning_rate": 1.7703352848054888e-06,
"loss": 0.3658,
"step": 420
},
{
"epoch": 1.9545454545454546,
"grad_norm": 2.4339048862457275,
"learning_rate": 1.6449496416858285e-06,
"loss": 0.3731,
"step": 430
},
{
"epoch": 2.0,
"grad_norm": 2.40981125831604,
"learning_rate": 1.521955206326976e-06,
"loss": 0.3567,
"step": 440
},
{
"epoch": 2.0454545454545454,
"grad_norm": 2.5268046855926514,
"learning_rate": 1.401695941216644e-06,
"loss": 0.2403,
"step": 450
},
{
"epoch": 2.090909090909091,
"grad_norm": 2.764014720916748,
"learning_rate": 1.2845081597488288e-06,
"loss": 0.2391,
"step": 460
},
{
"epoch": 2.1363636363636362,
"grad_norm": 2.833030939102173,
"learning_rate": 1.1707195857000218e-06,
"loss": 0.2451,
"step": 470
},
{
"epoch": 2.1818181818181817,
"grad_norm": 3.5294485092163086,
"learning_rate": 1.0606484367268906e-06,
"loss": 0.2646,
"step": 480
},
{
"epoch": 2.227272727272727,
"grad_norm": 2.8791749477386475,
"learning_rate": 9.546025344484868e-07,
"loss": 0.2316,
"step": 490
},
{
"epoch": 2.2727272727272725,
"grad_norm": 3.4600861072540283,
"learning_rate": 8.528784436016879e-07,
"loss": 0.2326,
"step": 500
},
{
"epoch": 2.3181818181818183,
"grad_norm": 3.8888542652130127,
"learning_rate": 7.557606426772962e-07,
"loss": 0.2425,
"step": 510
},
{
"epoch": 2.3636363636363638,
"grad_norm": 2.9734840393066406,
"learning_rate": 6.635207283561671e-07,
"loss": 0.2402,
"step": 520
},
{
"epoch": 2.409090909090909,
"grad_norm": 2.7985284328460693,
"learning_rate": 5.764166559702145e-07,
"loss": 0.2552,
"step": 530
},
{
"epoch": 2.4545454545454546,
"grad_norm": 2.636850595474243,
"learning_rate": 4.946920181123904e-07,
"loss": 0.2288,
"step": 540
},
{
"epoch": 2.5,
"grad_norm": 3.518472194671631,
"learning_rate": 4.185753634130718e-07,
"loss": 0.2302,
"step": 550
},
{
"epoch": 2.5454545454545454,
"grad_norm": 2.759793281555176,
"learning_rate": 3.482795573879241e-07,
"loss": 0.2486,
"step": 560
},
{
"epoch": 2.590909090909091,
"grad_norm": 3.072930335998535,
"learning_rate": 2.840011871446963e-07,
"loss": 0.2389,
"step": 570
},
{
"epoch": 2.6363636363636362,
"grad_norm": 2.546269178390503,
"learning_rate": 2.2592001161370392e-07,
"loss": 0.2463,
"step": 580
},
{
"epoch": 2.6818181818181817,
"grad_norm": 2.247389793395996,
"learning_rate": 1.74198458839491e-07,
"loss": 0.2235,
"step": 590
},
{
"epoch": 2.7272727272727275,
"grad_norm": 3.309213161468506,
"learning_rate": 1.289811717395087e-07,
"loss": 0.2358,
"step": 600
},
{
"epoch": 2.7727272727272725,
"grad_norm": 2.962289333343506,
"learning_rate": 9.03946036001449e-08,
"loss": 0.23,
"step": 610
},
{
"epoch": 2.8181818181818183,
"grad_norm": 2.887887954711914,
"learning_rate": 5.8546664441319346e-08,
"loss": 0.2348,
"step": 620
},
{
"epoch": 2.8636363636363638,
"grad_norm": 3.686352252960205,
"learning_rate": 3.352641923861144e-08,
"loss": 0.247,
"step": 630
},
{
"epoch": 2.909090909090909,
"grad_norm": 3.1452362537384033,
"learning_rate": 1.5403838846864694e-08,
"loss": 0.2284,
"step": 640
},
{
"epoch": 2.9545454545454546,
"grad_norm": 3.098659038543701,
"learning_rate": 4.229604321829561e-09,
"loss": 0.2312,
"step": 650
},
{
"epoch": 3.0,
"grad_norm": 2.993980884552002,
"learning_rate": 3.496518706597174e-11,
"loss": 0.2323,
"step": 660
},
{
"epoch": 3.0,
"step": 660,
"total_flos": 6.988388070252872e+17,
"train_loss": 0.41709409590923424,
"train_runtime": 37712.5646,
"train_samples_per_second": 0.14,
"train_steps_per_second": 0.018
}
],
"logging_steps": 10,
"max_steps": 660,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 1,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 6.988388070252872e+17,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink