Files changed (1) hide show
  1. README.md +4 -1
README.md CHANGED
@@ -18,7 +18,7 @@ Uncensored refers to the [ehartford/wizard_vicuna_70k_unfiltered](https://huggin
18
 
19
  Special thanks to [George Sung](https://huggingface.co/georgesung) for creating [llama2_7b_chat_uncensored](https://huggingface.co/georgesung/llama2_7b_chat_uncensored), and to [Eric Hartford](https://huggingface.co/ehartford/) for creating [ehartford/wizard_vicuna_70k_unfiltered](https://huggingface.co/datasets/ehartford/wizard_vicuna_70k_unfiltered)
20
 
21
- The version here is the fp16 HuggingFace model.
22
 
23
  In 8 bit mode, the model fits into 84% of A100 80GB (67.2GB) 68747MiB
24
  In 4 bit mode, the model fits into 51% of A100 80GB (40.8GB) 41559MiB
@@ -29,6 +29,9 @@ Thanks to [TheBloke](https://huggingface.co/TheBloke), he has created the GGML a
29
  * https://huggingface.co/TheBloke/llama2_70b_chat_uncensored-GGML
30
  * https://huggingface.co/TheBloke/llama2_70b_chat_uncensored-GPTQ
31
 
 
 
 
32
  # Prompt style
33
  The model was trained with the following prompt style:
34
  ```
 
18
 
19
  Special thanks to [George Sung](https://huggingface.co/georgesung) for creating [llama2_7b_chat_uncensored](https://huggingface.co/georgesung/llama2_7b_chat_uncensored), and to [Eric Hartford](https://huggingface.co/ehartford/) for creating [ehartford/wizard_vicuna_70k_unfiltered](https://huggingface.co/datasets/ehartford/wizard_vicuna_70k_unfiltered)
20
 
21
+ The version here is the fp32 HuggingFace model.
22
 
23
  In 8 bit mode, the model fits into 84% of A100 80GB (67.2GB) 68747MiB
24
  In 4 bit mode, the model fits into 51% of A100 80GB (40.8GB) 41559MiB
 
29
  * https://huggingface.co/TheBloke/llama2_70b_chat_uncensored-GGML
30
  * https://huggingface.co/TheBloke/llama2_70b_chat_uncensored-GPTQ
31
 
32
+ A smaller fp16 copy can be found at:
33
+ * https://huggingface.co/YokaiKoibito/llama2_70b_chat_uncensored
34
+
35
  # Prompt style
36
  The model was trained with the following prompt style:
37
  ```