Upload folder using huggingface_hub
Browse files
README.md
CHANGED
|
@@ -2,6 +2,23 @@
|
|
| 2 |
library_name: transformers
|
| 3 |
license: llama3
|
| 4 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
# Smaug-Llama-3-70B-Instruct-abliterated-v3 Model Card
|
| 6 |
|
| 7 |
[My Jupyter "cookbook" to replicate the methodology can be found here, refined library coming soon](https://huggingface.co/failspy/llama-3-70B-Instruct-abliterated/blob/main/ortho_cookbook.ipynb)
|
|
|
|
| 2 |
library_name: transformers
|
| 3 |
license: llama3
|
| 4 |
---
|
| 5 |
+
**Exllamav2** quant (**exl2** / **3.0 bpw**) made with ExLlamaV2 v0.0.21
|
| 6 |
+
|
| 7 |
+
Other EXL2 quants:
|
| 8 |
+
| **Quant** | **Model Size** | **lm_head** |
|
| 9 |
+
| ----- | ---------- | ------- |
|
| 10 |
+
|<center>**[2.2](https://huggingface.co/Zoyd/failspy_Smaug-Llama-3-70B-Instruct-abliterated-v3-2_2bpw_exl2)**</center> | <center>20886 MB</center> | <center>6</center> |
|
| 11 |
+
|<center>**[2.5](https://huggingface.co/Zoyd/failspy_Smaug-Llama-3-70B-Instruct-abliterated-v3-2_5bpw_exl2)**</center> | <center>23200 MB</center> | <center>6</center> |
|
| 12 |
+
|<center>**[3.0](https://huggingface.co/Zoyd/failspy_Smaug-Llama-3-70B-Instruct-abliterated-v3-3_0bpw_exl2)**</center> | <center>27269 MB</center> | <center>6</center> |
|
| 13 |
+
|<center>**[3.5](https://huggingface.co/Zoyd/failspy_Smaug-Llama-3-70B-Instruct-abliterated-v3-3_5bpw_exl2)**</center> | <center>31359 MB</center> | <center>6</center> |
|
| 14 |
+
|<center>**[3.75](https://huggingface.co/Zoyd/failspy_Smaug-Llama-3-70B-Instruct-abliterated-v3-3_75bpw_exl2)**</center> | <center>33395 MB</center> | <center>6</center> |
|
| 15 |
+
|<center>**[4.0](https://huggingface.co/Zoyd/failspy_Smaug-Llama-3-70B-Instruct-abliterated-v3-4_0bpw_exl2)**</center> | <center>35426 MB</center> | <center>6</center> |
|
| 16 |
+
|<center>**[4.25](https://huggingface.co/Zoyd/failspy_Smaug-Llama-3-70B-Instruct-abliterated-v3-4_25bpw_exl2)**</center> | <center>37478 MB</center> | <center>6</center> |
|
| 17 |
+
|<center>**[5.0](https://huggingface.co/Zoyd/failspy_Smaug-Llama-3-70B-Instruct-abliterated-v3-5_0bpw_exl2)**</center> | <center>43559 MB</center> | <center>6</center> |
|
| 18 |
+
|<center>**[6.0](https://huggingface.co/Zoyd/failspy_Smaug-Llama-3-70B-Instruct-abliterated-v3-6_0bpw_exl2)**</center> | <center>51958 MB</center> | <center>8</center> |
|
| 19 |
+
|<center>**[6.5](https://huggingface.co/Zoyd/failspy_Smaug-Llama-3-70B-Instruct-abliterated-v3-6_5bpw_exl2)**</center> | <center>56019 MB</center> | <center>8</center> |
|
| 20 |
+
|<center>**[8.0](https://huggingface.co/Zoyd/failspy_Smaug-Llama-3-70B-Instruct-abliterated-v3-8_0bpw_exl2)**</center> | <center>61865 MB</center> | <center>8</center> |
|
| 21 |
+
|
| 22 |
# Smaug-Llama-3-70B-Instruct-abliterated-v3 Model Card
|
| 23 |
|
| 24 |
[My Jupyter "cookbook" to replicate the methodology can be found here, refined library coming soon](https://huggingface.co/failspy/llama-3-70B-Instruct-abliterated/blob/main/ortho_cookbook.ipynb)
|