groxaxo/octen-embedding-8b-w4a16
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: groxaxo/octen-embedding-8b-w4a16
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-08-21 08:35:12 +08:00
commit 80d1d43641
13 changed files with 1248 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

15
validation-8b-auto-round-auto-gptq.json Normal file
View File

@@ -0,0 +1,15 @@
{
"base_model": "Octen/Octen-Embedding-8B",
"quant_model": "quantized/octen-embedding-8b-w4a16/auto-round-auto-gptq",
"num_eval_rows": 5,
"base_embedding_dim": 4096,
"quant_embedding_dim": 4096,
"base_recall_at_1": 0.8,
"base_recall_at_5": 1.0,
"quant_recall_at_1": 1.0,
"quant_recall_at_5": 1.0,
"mean_query_cosine_base_vs_quant": 0.9839715957641602,
"mean_doc_cosine_base_vs_quant": 0.9820238351821899,
"recall_at_1_delta": 0.19999999999999996,
"recall_at_5_delta": 0.0
}