Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,44 @@
|
|
1 |
---
|
2 |
license: apache-2.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: apache-2.0
|
3 |
+
pipeline_tag: text-generation
|
4 |
+
language:
|
5 |
+
- en
|
6 |
+
- he
|
7 |
+
tags:
|
8 |
+
- pretrained
|
9 |
+
inference:
|
10 |
+
parameters:
|
11 |
+
temperature: 0.7
|
12 |
---
|
13 |
+
|
14 |
+
[<img src="dicta-logo.jpg" width="300px"/>](https://dicta.org.il)
|
15 |
+
|
16 |
+
# Model Card for DictaLM-2.0
|
17 |
+
|
18 |
+
The DictaLM-2.0 Large Language Model (LLM) is a pretrained generative text model with 7 billion parameters trained to specialize in Hebrew text.
|
19 |
+
|
20 |
+
For full details of this model please read our [release blog post](https://dicta.org.il/dicta-lm).
|
21 |
+
|
22 |
+
This is the base model designed for completion (not for chat!) in the GGUF format for use with llama.cpp.
|
23 |
+
|
24 |
+
There are two versions available - float16 precision (`*.F16.gguf`) and 4-bit quantized precision (`*.Q4_K_M.gguf`).
|
25 |
+
|
26 |
+
You can view and access the full collection of base/instruct unquantized/quantized versions of `DictaLM-2.0` [here](https://huggingface.co/collections/dicta-il/dicta-lm-20-collection-661bbda397df671e4a430c27).
|
27 |
+
|
28 |
+
## Model Architecture
|
29 |
+
|
30 |
+
DictaLM-2.0 is based on the [Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) model with the following changes:
|
31 |
+
- An extended tokenizer with 1,000 injected tokens specifically for Hebrew, increasing the compression rate from 5.78 tokens/word to 2.76 tokens/word.
|
32 |
+
- Continued pretraining on over 190B tokens of naturally occuring text, 50% Hebrew and 50% English.
|
33 |
+
|
34 |
+
## Notice
|
35 |
+
|
36 |
+
DictaLM 2.0 is a pretrained base model and therefore does not have any moderation mechanisms.
|
37 |
+
|
38 |
+
## Citation
|
39 |
+
|
40 |
+
If you use this model, please cite:
|
41 |
+
|
42 |
+
```bibtex
|
43 |
+
[Will be added soon]
|
44 |
+
```
|