Commit
•
a8185f7
1
Parent(s):
1dba9fc
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
language:
|
4 |
+
- en
|
5 |
+
---
|
6 |
+
# Model Card for Model ID
|
7 |
+
|
8 |
+
This is Meta's Llama 2 7B quantized in 2-bit using AutoGPTQ from Hugging Face Transformers.
|
9 |
+
## Model Details
|
10 |
+
|
11 |
+
### Model Description
|
12 |
+
|
13 |
+
<!-- Provide a longer summary of what this model is. -->
|
14 |
+
|
15 |
+
|
16 |
+
|
17 |
+
- **Developed by:** [The Kaitchup](https://kaitchup.substack.com/)
|
18 |
+
- **Model type:** Causal (Llama 2)
|
19 |
+
- **Language(s) (NLP):** English
|
20 |
+
- **License:** [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0), [Llama 2 license agreement](https://ai.meta.com/resources/models-and-libraries/llama-downloads/)
|
21 |
+
|
22 |
+
### Model Sources
|
23 |
+
|
24 |
+
The method and code used to quantize the model are explained here:
|
25 |
+
[Quantize and Fine-tune LLMs with GPTQ Using Transformers and TRL](https://kaitchup.substack.com/p/quantize-and-fine-tune-llms-with)
|
26 |
+
|
27 |
+
## Uses
|
28 |
+
|
29 |
+
This model is pre-trained and not fine-tuned. You may fine-tune it with PEFT using adapters.
|
30 |
+
Note that the 2-bit quantization significantly decreases the performance of Llama 2.
|
31 |
+
|
32 |
+
|
33 |
+
## Other versions
|
34 |
+
|
35 |
+
- [kaitchup/Llama-2-7b-gptq-4bit](https://huggingface.co/kaitchup/Llama-2-7b-gptq-4bit)
|
36 |
+
- [kaitchup/Llama-2-7b-gptq-3bit](https://huggingface.co/kaitchup/Llama-2-7b-gptq-3bit)
|
37 |
+
|
38 |
+
|
39 |
+
|
40 |
+
|
41 |
+
## Model Card Contact
|
42 |
+
|
43 |
+
[The Kaitchup](https://kaitchup.substack.com/)
|
44 |
+
|
45 |
+
|