laion/nemotron-terminal-corpus-unified-3160__Qwen3-8B
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: laion/nemotron-terminal-corpus-unified-3160__Qwen3-8B
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-05-28 00:35:47 +08:00
commit 6e6b80a669
23 changed files with 153017 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

549
trainer_state.json Normal file
View File

@@ -0,0 +1,549 @@
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 7.0,
"eval_steps": 500,
"global_step": 231,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.15151515151515152,
"grad_norm": 8.12238366792196,
"learning_rate": 6.666666666666667e-06,
"loss": 0.9445,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.3465409278869629,
"step": 5,
"valid_targets_mean": 12230.9,
"valid_targets_min": 2381
},
{
"epoch": 0.30303030303030304,
"grad_norm": 2.3244763712891396,
"learning_rate": 1.5000000000000002e-05,
"loss": 0.8623,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.26335418224334717,
"step": 10,
"valid_targets_mean": 9716.3,
"valid_targets_min": 2066
},
{
"epoch": 0.45454545454545453,
"grad_norm": 1.6338159924911353,
"learning_rate": 2.3333333333333336e-05,
"loss": 0.7769,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.24279282987117767,
"step": 15,
"valid_targets_mean": 9685.4,
"valid_targets_min": 1674
},
{
"epoch": 0.6060606060606061,
"grad_norm": 0.855409558379507,
"learning_rate": 3.1666666666666666e-05,
"loss": 0.7316,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.2125861495733261,
"step": 20,
"valid_targets_mean": 9719.6,
"valid_targets_min": 4025
},
{
"epoch": 0.7575757575757576,
"grad_norm": 0.6878996427884594,
"learning_rate": 4e-05,
"loss": 0.6956,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.22667255997657776,
"step": 25,
"valid_targets_mean": 11508.9,
"valid_targets_min": 1411
},
{
"epoch": 0.9090909090909091,
"grad_norm": 0.5109614726377141,
"learning_rate": 3.994244399375679e-05,
"loss": 0.6737,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.1945917308330536,
"step": 30,
"valid_targets_mean": 9065.9,
"valid_targets_min": 2877
},
{
"epoch": 1.0606060606060606,
"grad_norm": 0.3600038057004485,
"learning_rate": 3.977010724441261e-05,
"loss": 0.6398,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.25795793533325195,
"step": 35,
"valid_targets_mean": 13362.0,
"valid_targets_min": 1740
},
{
"epoch": 1.2121212121212122,
"grad_norm": 0.31227917597174726,
"learning_rate": 3.9483981653469586e-05,
"loss": 0.6208,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.21348276734352112,
"step": 40,
"valid_targets_mean": 11814.1,
"valid_targets_min": 3326
},
{
"epoch": 1.3636363636363638,
"grad_norm": 0.3090682310426614,
"learning_rate": 3.908571404555758e-05,
"loss": 0.6038,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.1507582664489746,
"step": 45,
"valid_targets_mean": 7689.3,
"valid_targets_min": 1669
},
{
"epoch": 1.5151515151515151,
"grad_norm": 0.30361282072388873,
"learning_rate": 3.8577596689969346e-05,
"loss": 0.5883,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.18516400456428528,
"step": 50,
"valid_targets_mean": 9284.1,
"valid_targets_min": 1048
},
{
"epoch": 1.6666666666666665,
"grad_norm": 0.22551660139743698,
"learning_rate": 3.7962554107273926e-05,
"loss": 0.5872,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.19901783764362335,
"step": 55,
"valid_targets_mean": 10446.1,
"valid_targets_min": 1770
},
{
"epoch": 1.8181818181818183,
"grad_norm": 0.23298985586480123,
"learning_rate": 3.724412623694427e-05,
"loss": 0.5734,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.18426942825317383,
"step": 60,
"valid_targets_mean": 10647.8,
"valid_targets_min": 2295
},
{
"epoch": 1.9696969696969697,
"grad_norm": 0.2303711147417891,
"learning_rate": 3.642644806287938e-05,
"loss": 0.5755,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.18550506234169006,
"step": 65,
"valid_targets_mean": 10255.5,
"valid_targets_min": 2301
},
{
"epoch": 2.121212121212121,
"grad_norm": 0.22395388452993772,
"learning_rate": 3.55142258140884e-05,
"loss": 0.5548,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.1839693784713745,
"step": 70,
"valid_targets_mean": 10588.8,
"valid_targets_min": 2757
},
{
"epoch": 2.2727272727272725,
"grad_norm": 0.2742150018494869,
"learning_rate": 3.451270987751598e-05,
"loss": 0.5656,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.1844964623451233,
"step": 75,
"valid_targets_mean": 10316.0,
"valid_targets_min": 2118
},
{
"epoch": 2.4242424242424243,
"grad_norm": 0.23481939588137252,
"learning_rate": 3.342766457891194e-05,
"loss": 0.5443,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.17283733189105988,
"step": 80,
"valid_targets_mean": 10571.7,
"valid_targets_min": 2248
},
{
"epoch": 2.5757575757575757,
"grad_norm": 0.22913994808121177,
"learning_rate": 3.226533500567433e-05,
"loss": 0.5375,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.16651670634746552,
"step": 85,
"valid_targets_mean": 9745.1,
"valid_targets_min": 1978
},
{
"epoch": 2.7272727272727275,
"grad_norm": 0.24302031326379242,
"learning_rate": 3.1032411062620544e-05,
"loss": 0.5501,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.19409911334514618,
"step": 90,
"valid_targets_mean": 11224.8,
"valid_targets_min": 1389
},
{
"epoch": 2.878787878787879,
"grad_norm": 0.23049710507215065,
"learning_rate": 2.973598896756697e-05,
"loss": 0.5323,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.18646439909934998,
"step": 95,
"valid_targets_mean": 11390.2,
"valid_targets_min": 2219
},
{
"epoch": 3.0303030303030303,
"grad_norm": 0.214839079028194,
"learning_rate": 2.8383530408333285e-05,
"loss": 0.53,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.1786148101091385,
"step": 100,
"valid_targets_mean": 11495.4,
"valid_targets_min": 2884
},
{
"epoch": 3.1818181818181817,
"grad_norm": 0.2254674378323844,
"learning_rate": 2.6982819596247373e-05,
"loss": 0.525,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.15925350785255432,
"step": 105,
"valid_targets_mean": 9950.5,
"valid_targets_min": 1751
},
{
"epoch": 3.3333333333333335,
"grad_norm": 0.2196408033069726,
"learning_rate": 2.554191846333378e-05,
"loss": 0.5309,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.18058490753173828,
"step": 110,
"valid_targets_mean": 12094.2,
"valid_targets_min": 1840
},
{
"epoch": 3.484848484848485,
"grad_norm": 0.23943747979649102,
"learning_rate": 2.4069120261052682e-05,
"loss": 0.5219,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.19500833749771118,
"step": 115,
"valid_targets_mean": 12582.8,
"valid_targets_min": 2484
},
{
"epoch": 3.6363636363636362,
"grad_norm": 0.2348454371321129,
"learning_rate": 2.2572901827656626e-05,
"loss": 0.5134,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.1679806411266327,
"step": 120,
"valid_targets_mean": 9697.8,
"valid_targets_min": 1978
},
{
"epoch": 3.787878787878788,
"grad_norm": 0.2246419072884847,
"learning_rate": 2.1061874798894992e-05,
"loss": 0.5222,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.16158881783485413,
"step": 125,
"valid_targets_mean": 10328.7,
"valid_targets_min": 1344
},
{
"epoch": 3.9393939393939394,
"grad_norm": 0.2355398761218515,
"learning_rate": 1.9544736042877886e-05,
"loss": 0.523,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.16300621628761292,
"step": 130,
"valid_targets_mean": 9793.8,
"valid_targets_min": 1479
},
{
"epoch": 4.090909090909091,
"grad_norm": 0.22067466285967524,
"learning_rate": 1.8030217604376628e-05,
"loss": 0.5136,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.18298637866973877,
"step": 135,
"valid_targets_mean": 10920.6,
"valid_targets_min": 2219
},
{
"epoch": 4.242424242424242,
"grad_norm": 0.20743964532905532,
"learning_rate": 1.6527036446661396e-05,
"loss": 0.5168,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.19334109127521515,
"step": 140,
"valid_targets_mean": 11248.1,
"valid_targets_min": 1247
},
{
"epoch": 4.393939393939394,
"grad_norm": 0.23476266334929202,
"learning_rate": 1.5043844280142005e-05,
"loss": 0.5167,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.14934216439723969,
"step": 145,
"valid_targets_mean": 8360.1,
"valid_targets_min": 2467
},
{
"epoch": 4.545454545454545,
"grad_norm": 0.19627736108386315,
"learning_rate": 1.358917776657806e-05,
"loss": 0.5173,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.18021146953105927,
"step": 150,
"valid_targets_mean": 12818.7,
"valid_targets_min": 2883
},
{
"epoch": 4.696969696969697,
"grad_norm": 0.21398920361472892,
"learning_rate": 1.2171409385463218e-05,
"loss": 0.5031,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.16597239673137665,
"step": 155,
"valid_targets_mean": 10478.3,
"valid_targets_min": 1655
},
{
"epoch": 4.848484848484849,
"grad_norm": 0.21179990608393784,
"learning_rate": 1.0798699245376959e-05,
"loss": 0.511,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.1842915564775467,
"step": 160,
"valid_targets_mean": 10866.0,
"valid_targets_min": 2629
},
{
"epoch": 5.0,
"grad_norm": 0.20243539651963496,
"learning_rate": 9.478948117658577e-06,
"loss": 0.5073,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.1693403720855713,
"step": 165,
"valid_targets_mean": 11555.1,
"valid_targets_min": 2400
},
{
"epoch": 5.151515151515151,
"grad_norm": 0.21995234700338384,
"learning_rate": 8.219751962722726e-06,
"loss": 0.5009,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.19458279013633728,
"step": 170,
"valid_targets_mean": 10903.9,
"valid_targets_min": 2912
},
{
"epoch": 5.303030303030303,
"grad_norm": 0.19961404869931895,
"learning_rate": 7.028358210744881e-06,
"loss": 0.5062,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.1704961061477661,
"step": 175,
"valid_targets_mean": 10323.8,
"valid_targets_min": 1674
},
{
"epoch": 5.454545454545454,
"grad_norm": 0.1868367050248924,
"learning_rate": 5.911624048347757e-06,
"loss": 0.5133,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.13904976844787598,
"step": 180,
"valid_targets_mean": 9239.8,
"valid_targets_min": 2227
},
{
"epoch": 5.606060606060606,
"grad_norm": 0.20255499362596352,
"learning_rate": 4.875976951373633e-06,
"loss": 0.4969,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.1814340353012085,
"step": 185,
"valid_targets_mean": 12320.0,
"valid_targets_min": 2401
},
{
"epoch": 5.757575757575758,
"grad_norm": 0.19634205407139418,
"learning_rate": 3.927377690900436e-06,
"loss": 0.5076,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.20642095804214478,
"step": 190,
"valid_targets_mean": 14353.4,
"valid_targets_min": 2651
},
{
"epoch": 5.909090909090909,
"grad_norm": 0.1764494951381787,
"learning_rate": 3.071286025423983e-06,
"loss": 0.5152,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.17783287167549133,
"step": 195,
"valid_targets_mean": 11610.0,
"valid_targets_min": 3112
},
{
"epoch": 6.0606060606060606,
"grad_norm": 0.1937790180719643,
"learning_rate": 2.312629276668554e-06,
"loss": 0.4961,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.12838870286941528,
"step": 200,
"valid_targets_mean": 7998.2,
"valid_targets_min": 2053
},
{
"epoch": 6.212121212121212,
"grad_norm": 0.19638286518502554,
"learning_rate": 1.6557739698909436e-06,
"loss": 0.5032,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.1665273904800415,
"step": 205,
"valid_targets_mean": 11573.5,
"valid_targets_min": 2883
},
{
"epoch": 6.363636363636363,
"grad_norm": 0.19162317469724846,
"learning_rate": 1.1045007019049182e-06,
"loss": 0.5088,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.1381627321243286,
"step": 210,
"valid_targets_mean": 9245.8,
"valid_targets_min": 2000
},
{
"epoch": 6.515151515151516,
"grad_norm": 0.19041257044671683,
"learning_rate": 6.619823814758786e-07,
"loss": 0.5082,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.17835193872451782,
"step": 215,
"valid_targets_mean": 10446.8,
"valid_targets_min": 2301
},
{
"epoch": 6.666666666666667,
"grad_norm": 0.1809565932702134,
"learning_rate": 3.307659673251595e-07,
"loss": 0.5039,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.1760205328464508,
"step": 220,
"valid_targets_mean": 11383.7,
"valid_targets_min": 2378
},
{
"epoch": 6.818181818181818,
"grad_norm": 0.18418769422837922,
"learning_rate": 1.1275780885282806e-07,
"loss": 0.5051,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.156514510512352,
"step": 225,
"valid_targets_mean": 9719.4,
"valid_targets_min": 2215
},
{
"epoch": 6.96969696969697,
"grad_norm": 0.199244385075515,
"learning_rate": 9.212673951897177e-09,
"loss": 0.5035,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.16282179951667786,
"step": 230,
"valid_targets_mean": 10562.4,
"valid_targets_min": 2828
},
{
"epoch": 7.0,
"step": 231,
"total_flos": 9.111210907244954e+17,
"train_loss": 0.0,
"train_runtime": 1.5616,
"train_samples_per_second": 14164.97,
"train_steps_per_second": 147.925
}
],
"logging_steps": 5,
"max_steps": 231,
"num_input_tokens_seen": 0,
"num_train_epochs": 7,
"save_steps": 300,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 9.111210907244954e+17,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}