Update README.md
Browse files
README.md
CHANGED
@@ -23,8 +23,9 @@ widget:
|
|
23 |
- source_sentence: Cats usually hate dogs.
|
24 |
sentences:
|
25 |
- Куда вы ходили в прошлое воскресенье?
|
26 |
-
-
|
27 |
-
|
|
|
28 |
- Mir tut der Arm weh.
|
29 |
- source_sentence: How foolish I was not to discover that simple lie!
|
30 |
sentences:
|
@@ -36,6 +37,7 @@ widget:
|
|
36 |
- Το σχολείο μας έχει εννιά τάξεις.
|
37 |
- When applying to American universities, your TOEFL score is only one factor.
|
38 |
- Je n'ai pas encore pris ma décision.
|
|
|
39 |
---
|
40 |
|
41 |
# SentenceTransformer based on agentlans/multilingual-e5-small-aligned
|
@@ -43,7 +45,8 @@ widget:
|
|
43 |
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [agentlans/multilingual-e5-small-aligned](https://huggingface.co/agentlans/multilingual-e5-small-aligned). It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
|
44 |
|
45 |
- One of the smallest multilingual embedding models on Huggingface
|
46 |
-
-
|
|
|
47 |
- Includes pairs where one or both sentences are non-English
|
48 |
- For each pair, two negative examples were generated
|
49 |
|
|
|
23 |
- source_sentence: Cats usually hate dogs.
|
24 |
sentences:
|
25 |
- Куда вы ходили в прошлое воскресенье?
|
26 |
+
- >-
|
27 |
+
The bottles of beer that I brought to the party were redundant; the host's
|
28 |
+
family owned a brewery.
|
29 |
- Mir tut der Arm weh.
|
30 |
- source_sentence: How foolish I was not to discover that simple lie!
|
31 |
sentences:
|
|
|
37 |
- Το σχολείο μας έχει εννιά τάξεις.
|
38 |
- When applying to American universities, your TOEFL score is only one factor.
|
39 |
- Je n'ai pas encore pris ma décision.
|
40 |
+
license: mit
|
41 |
---
|
42 |
|
43 |
# SentenceTransformer based on agentlans/multilingual-e5-small-aligned
|
|
|
45 |
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [agentlans/multilingual-e5-small-aligned](https://huggingface.co/agentlans/multilingual-e5-small-aligned). It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
|
46 |
|
47 |
- One of the smallest multilingual embedding models on Huggingface
|
48 |
+
- This model is aligned which means translations have similar embeddings compared to unrelated sentences
|
49 |
+
- Finetuned on 1,000,000 randomly selected sentence pairs downloaded from Tatoeba 2024-09-26
|
50 |
- Includes pairs where one or both sentences are non-English
|
51 |
- For each pair, two negative examples were generated
|
52 |
|