Update README.md
Browse files
README.md
CHANGED
|
@@ -25,6 +25,8 @@ The model produces **384-dimensional** embeddings (CLS token) optimized for cosi
|
|
| 25 |
|
| 26 |
* ~22 417 920M parameters
|
| 27 |
|
|
|
|
|
|
|
| 28 |
* **Max sequence length: 128 tokens**
|
| 29 |
|
| 30 |
* Loss: MultipleNegativesSymmetricRankingLoss (InfoNCE)
|
|
@@ -35,6 +37,8 @@ The model produces **384-dimensional** embeddings (CLS token) optimized for cosi
|
|
| 35 |
|
| 36 |
* Data: ~500k triplets from MS MARCO
|
| 37 |
|
|
|
|
|
|
|
| 38 |
---
|
| 39 |
|
| 40 |
# Evaluation
|
|
|
|
| 25 |
|
| 26 |
* ~22 417 920M parameters
|
| 27 |
|
| 28 |
+
* Vocab Size: 30 522
|
| 29 |
+
|
| 30 |
* **Max sequence length: 128 tokens**
|
| 31 |
|
| 32 |
* Loss: MultipleNegativesSymmetricRankingLoss (InfoNCE)
|
|
|
|
| 37 |
|
| 38 |
* Data: ~500k triplets from MS MARCO
|
| 39 |
|
| 40 |
+
|
| 41 |
+
|
| 42 |
---
|
| 43 |
|
| 44 |
# Evaluation
|