UnstableLlama commited on
Commit
a266222
1 Parent(s): 74253ab

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +10 -0
README.md CHANGED
@@ -3,6 +3,16 @@ license: llama3.1
3
  ---
4
  # Rombos-LLM-V2.6-Nemotron-70b
5
 
 
 
 
 
 
 
 
 
 
 
6
  ![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/642cc1c253e76b4c2286c58e/lLAQ7FGTYDz0xT8rs9-ti.jpeg)
7
 
8
  I applied the last step of my continuous finetuning method to the Nemotron-70b model from Nvidia. More details bellow:
 
3
  ---
4
  # Rombos-LLM-V2.6-Nemotron-70b
5
 
6
+ ---
7
+ <p><h2>ExLlamaV2 Quantization</h2></p>
8
+ <p>Quantized with <a href="https://github.com/turboderp/exllamav2/releases/tag/v0.2.2">ExLlamaV2 v0.2.3</a></p>
9
+
10
+ [2.2 Bits Per Weight](https://huggingface.co/UnstableLlama/Rombos-LLM-V2.6-Nemotron-70b-exl2/tree/2_2)
11
+
12
+ [4.65 Bits Per Weight](https://huggingface.co/UnstableLlama/Rombos-LLM-V2.6-Nemotron-70b-exl2/tree/4_65)
13
+
14
+ ---
15
+
16
  ![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/642cc1c253e76b4c2286c58e/lLAQ7FGTYDz0xT8rs9-ti.jpeg)
17
 
18
  I applied the last step of my continuous finetuning method to the Nemotron-70b model from Nvidia. More details bellow: