laion/allenai-sera-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/allenai-sera-unified-3160__Qwen3-8B
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-05-06 06:22:47 +08:00
commit afe1b83817
23 changed files with 153020 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": 4.114635968171536,
"learning_rate": 6.666666666666667e-06,
"loss": 0.4313,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.13727591931819916,
"step": 5,
"valid_targets_mean": 14623.7,
"valid_targets_min": 8472
},
{
"epoch": 0.30303030303030304,
"grad_norm": 1.4536543569838813,
"learning_rate": 1.5000000000000002e-05,
"loss": 0.3895,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.12720003724098206,
"step": 10,
"valid_targets_mean": 13794.9,
"valid_targets_min": 8012
},
{
"epoch": 0.45454545454545453,
"grad_norm": 0.5050980157229342,
"learning_rate": 2.3333333333333336e-05,
"loss": 0.3418,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.10843911021947861,
"step": 15,
"valid_targets_mean": 13961.8,
"valid_targets_min": 7669
},
{
"epoch": 0.6060606060606061,
"grad_norm": 0.42246792506810416,
"learning_rate": 3.1666666666666666e-05,
"loss": 0.3153,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.10476820170879364,
"step": 20,
"valid_targets_mean": 13416.8,
"valid_targets_min": 8687
},
{
"epoch": 0.7575757575757576,
"grad_norm": 0.32959630386266386,
"learning_rate": 4e-05,
"loss": 0.2947,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.09988001734018326,
"step": 25,
"valid_targets_mean": 14083.4,
"valid_targets_min": 6271
},
{
"epoch": 0.9090909090909091,
"grad_norm": 0.2350625305435842,
"learning_rate": 3.994244399375679e-05,
"loss": 0.271,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.09274592995643616,
"step": 30,
"valid_targets_mean": 14132.5,
"valid_targets_min": 5776
},
{
"epoch": 1.0606060606060606,
"grad_norm": 0.19138295760535673,
"learning_rate": 3.977010724441261e-05,
"loss": 0.2523,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.08168983459472656,
"step": 35,
"valid_targets_mean": 13705.5,
"valid_targets_min": 6252
},
{
"epoch": 1.2121212121212122,
"grad_norm": 0.16875241063378532,
"learning_rate": 3.9483981653469586e-05,
"loss": 0.2413,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.07980272173881531,
"step": 40,
"valid_targets_mean": 13126.1,
"valid_targets_min": 3159
},
{
"epoch": 1.3636363636363638,
"grad_norm": 0.14569427919381664,
"learning_rate": 3.908571404555758e-05,
"loss": 0.2315,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.0792369619011879,
"step": 45,
"valid_targets_mean": 14597.3,
"valid_targets_min": 6419
},
{
"epoch": 1.5151515151515151,
"grad_norm": 0.12740714547686136,
"learning_rate": 3.8577596689969346e-05,
"loss": 0.228,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.07131356000900269,
"step": 50,
"valid_targets_mean": 14610.6,
"valid_targets_min": 9878
},
{
"epoch": 1.6666666666666665,
"grad_norm": 0.13219914885031364,
"learning_rate": 3.7962554107273926e-05,
"loss": 0.2184,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.07061326503753662,
"step": 55,
"valid_targets_mean": 14205.5,
"valid_targets_min": 8994
},
{
"epoch": 1.8181818181818183,
"grad_norm": 0.12537992825560595,
"learning_rate": 3.724412623694427e-05,
"loss": 0.2129,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.0706273764371872,
"step": 60,
"valid_targets_mean": 14081.0,
"valid_targets_min": 7128
},
{
"epoch": 1.9696969696969697,
"grad_norm": 0.1317216509879008,
"learning_rate": 3.642644806287938e-05,
"loss": 0.2075,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.06866224110126495,
"step": 65,
"valid_targets_mean": 13509.2,
"valid_targets_min": 6997
},
{
"epoch": 2.121212121212121,
"grad_norm": 0.13400162499874246,
"learning_rate": 3.55142258140884e-05,
"loss": 0.2062,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.06507541239261627,
"step": 70,
"valid_targets_mean": 13193.5,
"valid_targets_min": 3186
},
{
"epoch": 2.2727272727272725,
"grad_norm": 0.14105797671143652,
"learning_rate": 3.451270987751598e-05,
"loss": 0.2062,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.07145757973194122,
"step": 75,
"valid_targets_mean": 14511.4,
"valid_targets_min": 5311
},
{
"epoch": 2.4242424242424243,
"grad_norm": 0.1342111928769276,
"learning_rate": 3.342766457891194e-05,
"loss": 0.1972,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.07077564299106598,
"step": 80,
"valid_targets_mean": 14961.1,
"valid_targets_min": 9318
},
{
"epoch": 2.5757575757575757,
"grad_norm": 0.12636739349250312,
"learning_rate": 3.226533500567433e-05,
"loss": 0.1948,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.0638246163725853,
"step": 85,
"valid_targets_mean": 14331.9,
"valid_targets_min": 8629
},
{
"epoch": 2.7272727272727275,
"grad_norm": 0.1376354010348529,
"learning_rate": 3.1032411062620544e-05,
"loss": 0.1969,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.06627653539180756,
"step": 90,
"valid_targets_mean": 14391.3,
"valid_targets_min": 9017
},
{
"epoch": 2.878787878787879,
"grad_norm": 0.12491286274381265,
"learning_rate": 2.973598896756697e-05,
"loss": 0.1923,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.05493774637579918,
"step": 95,
"valid_targets_mean": 14151.3,
"valid_targets_min": 6828
},
{
"epoch": 3.0303030303030303,
"grad_norm": 0.13478052499964224,
"learning_rate": 2.8383530408333285e-05,
"loss": 0.1914,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.06786017119884491,
"step": 100,
"valid_targets_mean": 13563.6,
"valid_targets_min": 5230
},
{
"epoch": 3.1818181818181817,
"grad_norm": 0.13423564287671794,
"learning_rate": 2.6982819596247373e-05,
"loss": 0.1892,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.06517742574214935,
"step": 105,
"valid_targets_mean": 13841.5,
"valid_targets_min": 7737
},
{
"epoch": 3.3333333333333335,
"grad_norm": 0.12892135797363977,
"learning_rate": 2.554191846333378e-05,
"loss": 0.1852,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.06293950974941254,
"step": 110,
"valid_targets_mean": 13982.2,
"valid_targets_min": 6906
},
{
"epoch": 3.484848484848485,
"grad_norm": 0.13050995936742432,
"learning_rate": 2.4069120261052682e-05,
"loss": 0.1861,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.060917340219020844,
"step": 115,
"valid_targets_mean": 14595.4,
"valid_targets_min": 7001
},
{
"epoch": 3.6363636363636362,
"grad_norm": 0.12603585687316865,
"learning_rate": 2.2572901827656626e-05,
"loss": 0.1872,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.06335198879241943,
"step": 120,
"valid_targets_mean": 14134.8,
"valid_targets_min": 4539
},
{
"epoch": 3.787878787878788,
"grad_norm": 0.12458026081937879,
"learning_rate": 2.1061874798894992e-05,
"loss": 0.1849,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.0566466748714447,
"step": 125,
"valid_targets_mean": 13645.1,
"valid_targets_min": 6953
},
{
"epoch": 3.9393939393939394,
"grad_norm": 0.1237313936998869,
"learning_rate": 1.9544736042877886e-05,
"loss": 0.1827,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.06262873858213425,
"step": 130,
"valid_targets_mean": 14213.2,
"valid_targets_min": 7309
},
{
"epoch": 4.090909090909091,
"grad_norm": 0.12687157253526618,
"learning_rate": 1.8030217604376628e-05,
"loss": 0.183,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.06545929610729218,
"step": 135,
"valid_targets_mean": 14312.2,
"valid_targets_min": 4618
},
{
"epoch": 4.242424242424242,
"grad_norm": 0.13286557514006167,
"learning_rate": 1.6527036446661396e-05,
"loss": 0.1838,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.0621417798101902,
"step": 140,
"valid_targets_mean": 13995.9,
"valid_targets_min": 3159
},
{
"epoch": 4.393939393939394,
"grad_norm": 0.13911478169619562,
"learning_rate": 1.5043844280142005e-05,
"loss": 0.1811,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.06362389773130417,
"step": 145,
"valid_targets_mean": 14415.2,
"valid_targets_min": 7395
},
{
"epoch": 4.545454545454545,
"grad_norm": 0.12748545316526233,
"learning_rate": 1.358917776657806e-05,
"loss": 0.1782,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.05887322872877121,
"step": 150,
"valid_targets_mean": 13608.7,
"valid_targets_min": 4237
},
{
"epoch": 4.696969696969697,
"grad_norm": 0.12065267038653112,
"learning_rate": 1.2171409385463218e-05,
"loss": 0.1767,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.05912807583808899,
"step": 155,
"valid_targets_mean": 13554.9,
"valid_targets_min": 6402
},
{
"epoch": 4.848484848484849,
"grad_norm": 0.1242188167578349,
"learning_rate": 1.0798699245376959e-05,
"loss": 0.1781,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.05947456508874893,
"step": 160,
"valid_targets_mean": 14594.9,
"valid_targets_min": 7248
},
{
"epoch": 5.0,
"grad_norm": 0.14434343975971592,
"learning_rate": 9.478948117658577e-06,
"loss": 0.1786,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.059081535786390305,
"step": 165,
"valid_targets_mean": 13341.7,
"valid_targets_min": 4728
},
{
"epoch": 5.151515151515151,
"grad_norm": 0.13210237534940392,
"learning_rate": 8.219751962722726e-06,
"loss": 0.1803,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.059808894991874695,
"step": 170,
"valid_targets_mean": 14140.9,
"valid_targets_min": 5241
},
{
"epoch": 5.303030303030303,
"grad_norm": 0.11760754357701442,
"learning_rate": 7.028358210744881e-06,
"loss": 0.1723,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.05813150107860565,
"step": 175,
"valid_targets_mean": 13437.8,
"valid_targets_min": 3456
},
{
"epoch": 5.454545454545454,
"grad_norm": 0.23292822424304802,
"learning_rate": 5.911624048347757e-06,
"loss": 0.1782,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.054474152624607086,
"step": 180,
"valid_targets_mean": 13450.3,
"valid_targets_min": 8527
},
{
"epoch": 5.606060606060606,
"grad_norm": 0.11629211221337708,
"learning_rate": 4.875976951373633e-06,
"loss": 0.177,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.05472658574581146,
"step": 185,
"valid_targets_mean": 14527.5,
"valid_targets_min": 6291
},
{
"epoch": 5.757575757575758,
"grad_norm": 0.12053316478201255,
"learning_rate": 3.927377690900436e-06,
"loss": 0.1755,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.06293582916259766,
"step": 190,
"valid_targets_mean": 15079.5,
"valid_targets_min": 5311
},
{
"epoch": 5.909090909090909,
"grad_norm": 0.14093615075318672,
"learning_rate": 3.071286025423983e-06,
"loss": 0.1762,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.0564623698592186,
"step": 195,
"valid_targets_mean": 13432.3,
"valid_targets_min": 4391
},
{
"epoch": 6.0606060606060606,
"grad_norm": 0.11794615480974098,
"learning_rate": 2.312629276668554e-06,
"loss": 0.1746,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.059501372277736664,
"step": 200,
"valid_targets_mean": 14715.9,
"valid_targets_min": 8912
},
{
"epoch": 6.212121212121212,
"grad_norm": 0.12194911582381218,
"learning_rate": 1.6557739698909436e-06,
"loss": 0.1764,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.0586993545293808,
"step": 205,
"valid_targets_mean": 13863.8,
"valid_targets_min": 5335
},
{
"epoch": 6.363636363636363,
"grad_norm": 0.12137779369589421,
"learning_rate": 1.1045007019049182e-06,
"loss": 0.1753,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.05951303243637085,
"step": 210,
"valid_targets_mean": 13876.2,
"valid_targets_min": 4510
},
{
"epoch": 6.515151515151516,
"grad_norm": 0.11115005771790042,
"learning_rate": 6.619823814758786e-07,
"loss": 0.1725,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.0475098192691803,
"step": 215,
"valid_targets_mean": 13260.7,
"valid_targets_min": 2975
},
{
"epoch": 6.666666666666667,
"grad_norm": 0.11465648239113958,
"learning_rate": 3.307659673251595e-07,
"loss": 0.1759,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.05422068014740944,
"step": 220,
"valid_targets_mean": 13885.9,
"valid_targets_min": 4618
},
{
"epoch": 6.818181818181818,
"grad_norm": 0.12646402426260042,
"learning_rate": 1.1275780885282806e-07,
"loss": 0.1764,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.06054677814245224,
"step": 225,
"valid_targets_mean": 13998.0,
"valid_targets_min": 7669
},
{
"epoch": 6.96969696969697,
"grad_norm": 0.11699617410563377,
"learning_rate": 9.212673951897177e-09,
"loss": 0.176,
"loss_nan_ranks": 0,
"loss_rank_avg": 0.061230286955833435,
"step": 230,
"valid_targets_mean": 14301.2,
"valid_targets_min": 7992
},
{
"epoch": 7.0,
"step": 231,
"total_flos": 1.7948424939556045e+18,
"train_loss": 0.0,
"train_runtime": 1.1533,
"train_samples_per_second": 19179.804,
"train_steps_per_second": 200.295
}
],
"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": 1.7948424939556045e+18,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}