jionglin
commited on
Commit
·
ad225ed
1
Parent(s):
b0ddac9
commit modelcard
Browse files
README.md
CHANGED
|
@@ -1,4 +1,4 @@
|
|
| 1 |
-
|
| 2 |
tags:
|
| 3 |
- mteb
|
| 4 |
model-index:
|
|
@@ -1082,8 +1082,7 @@ model-index:
|
|
| 1082 |
value: 60.864524843400616
|
| 1083 |
- type: f1
|
| 1084 |
value: 79.41246877404483
|
| 1085 |
-
|
| 1086 |
-
```
|
| 1087 |
|
| 1088 |
ZNV Embedding utilizes a 6B LLM (Large Language Model) for embedding, achieving excellent embedding results.
|
| 1089 |
|
|
@@ -1203,5 +1202,4 @@ with torch.no_grad():
|
|
| 1203 |
output = znv_model(["请问你的电话号码是多少?","可以告诉我你的手机号吗?"])
|
| 1204 |
cos_sim = F.cosine_similarity(output[0],output[1],dim=0)
|
| 1205 |
print(cos_sim)
|
| 1206 |
-
```
|
| 1207 |
-
|
|
|
|
| 1 |
+
---
|
| 2 |
tags:
|
| 3 |
- mteb
|
| 4 |
model-index:
|
|
|
|
| 1082 |
value: 60.864524843400616
|
| 1083 |
- type: f1
|
| 1084 |
value: 79.41246877404483
|
| 1085 |
+
---
|
|
|
|
| 1086 |
|
| 1087 |
ZNV Embedding utilizes a 6B LLM (Large Language Model) for embedding, achieving excellent embedding results.
|
| 1088 |
|
|
|
|
| 1202 |
output = znv_model(["请问你的电话号码是多少?","可以告诉我你的手机号吗?"])
|
| 1203 |
cos_sim = F.cosine_similarity(output[0],output[1],dim=0)
|
| 1204 |
print(cos_sim)
|
| 1205 |
+
```
|
|
|