prithivida
commited on
Commit
•
adba986
1
Parent(s):
8ded640
Update README.md
Browse files
README.md
CHANGED
@@ -34,7 +34,8 @@ pipeline_tag: sentence-similarity
|
|
34 |
|
35 |
- Model: BERT.
|
36 |
- Tokenizer: XLM-Roberta's Tokenizer.
|
37 |
-
-
|
|
|
38 |
<br/>
|
39 |
|
40 |
<center>
|
@@ -212,7 +213,7 @@ Here are our numbers for the full hindi run on BGE-M3
|
|
212 |
{'MRR@10': 0.60893, 'MRR@100': 0.615, 'MRR@1000': 0.6151}
|
213 |
```
|
214 |
|
215 |
-
Fair warning BGE-M3 is $ expensive to evaluate, probably that's why it's not part of any of the retrieval slice of MTEB benchmarks.
|
216 |
|
217 |
|
218 |
# Reference:
|
|
|
34 |
|
35 |
- Model: BERT.
|
36 |
- Tokenizer: XLM-Roberta's Tokenizer.
|
37 |
+
- Vocab: 250K
|
38 |
+
|
39 |
<br/>
|
40 |
|
41 |
<center>
|
|
|
213 |
{'MRR@10': 0.60893, 'MRR@100': 0.615, 'MRR@1000': 0.6151}
|
214 |
```
|
215 |
|
216 |
+
Fair warning BGE-M3 is $ expensive to evaluate, probably* that's why it's not part of any of the retrieval slice of MTEB benchmarks.
|
217 |
|
218 |
|
219 |
# Reference:
|