aifeifei798 commited on
Commit
3fdc11b
1 Parent(s): 218834d

Upload README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -7
README.md CHANGED
@@ -14,11 +14,7 @@ tags:
14
  - Lewdiculous's superb gguf version, thank you for your conscientious and responsible dedication.
15
  - https://huggingface.co/LWDCLS/llama3-8B-DarkIdol-2.2-Uncensored-1048K-GGUF-IQ-Imatrix-Request
16
 
17
- # fast quantizations
18
- - The difference with normal quantizations is that I quantize the output and embed tensors to f16.and the other tensors to 15_k,q6_k or q8_0.This creates models that are little or not degraded at all and have a smaller size.They run at about 3-6 t/sec on CPU only using llama.cpp And obviously faster on computers with potent GPUs
19
- - https://huggingface.co/ZeroWw/llama3-8B-DarkIdol-2.2-Uncensored-1048K-GGUF
20
- - More models here: https://huggingface.co/RobertSinclair
21
-
22
  ## Why 1048K?
23
  Due to the optimization of the preferred model, its performance is excellent across the range of 2000-1048K. Personal usage scenarios, such as 8186, 32K, etc., are insufficient. My primary role involves managing virtual idol Twitter accounts and assisting with singing, etc. A good conversation can be very lengthy, and sometimes even 32K is not enough. Imagine having a heated chat with your virtual girlfriend, only for it to abruptly cut off—that feeling is too painful.
24
 
@@ -39,8 +35,8 @@ The module combination has been readjusted to better fulfill various roles and h
39
  - DarkIdol:Roles that you can imagine and those that you cannot imagine.
40
  - Roleplay
41
  - Specialized in various role-playing scenarios
42
- - more look at test role. (https://huggingface.co/aifeifei798/llama3-8B-DarkIdol-1.2/resolve/main/test)
43
- - more look at LM Studio presets (https://huggingface.co/aifeifei798/llama3-8B-DarkIdol-1.2/resolve/main/config-presets)
44
 
45
  ![image/png](https://huggingface.co/aifeifei798/llama3-8B-DarkIdol-2.2-Uncensored-1048K/resolve/main/llama3-8B-DarkIdol-2.2-Uncensored-1048K.png)
46
 
 
14
  - Lewdiculous's superb gguf version, thank you for your conscientious and responsible dedication.
15
  - https://huggingface.co/LWDCLS/llama3-8B-DarkIdol-2.2-Uncensored-1048K-GGUF-IQ-Imatrix-Request
16
 
17
+
 
 
 
 
18
  ## Why 1048K?
19
  Due to the optimization of the preferred model, its performance is excellent across the range of 2000-1048K. Personal usage scenarios, such as 8186, 32K, etc., are insufficient. My primary role involves managing virtual idol Twitter accounts and assisting with singing, etc. A good conversation can be very lengthy, and sometimes even 32K is not enough. Imagine having a heated chat with your virtual girlfriend, only for it to abruptly cut off—that feeling is too painful.
20
 
 
35
  - DarkIdol:Roles that you can imagine and those that you cannot imagine.
36
  - Roleplay
37
  - Specialized in various role-playing scenarios
38
+ - more look at test role. (https://huggingface.co/aifeifei798/llama3-8B-DarkIdol-1.2/tree/main/test)
39
+ - more look at LM Studio presets (https://huggingface.co/aifeifei798/llama3-8B-DarkIdol-1.2/tree/main/config-presets)
40
 
41
  ![image/png](https://huggingface.co/aifeifei798/llama3-8B-DarkIdol-2.2-Uncensored-1048K/resolve/main/llama3-8B-DarkIdol-2.2-Uncensored-1048K.png)
42