Hugging Face's logo

lemonilia
/
ShoriRP-v0.75d

Text Generation
Transformers
GGUF
English
mistral
Not-For-All-Audiences
Inference Endpoints
4-bit precision
bitsandbytes
conversational
Model card Files Community
6
lemonilia commited on
Commit
93b3aa9
·
verified ·
1 Parent(s): c7ebab6

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -0
README.md CHANGED
@@ -44,6 +44,8 @@ Suitable `json` files have been provided to easily apply the prompting format in
44
  - [Context](https://huggingface.co/lemonilia/test-model-srp-0.5/blob/main/BlockML-Context.json)
45
  - [Instruct](https://huggingface.co/lemonilia/test-model-srp-0.5/resolve/main/BlockML-Instruct.json)
46
 
 
 
47
  ## Block characters
48
  The model uses a _ChatML-like_ prompting format with a few changes from the usual roles typically used for ChatGPT-like assistant chatbots. The first is that to avoid complications with tokenization and special tokens, `<|im_start|>` has been replaced with `▀` (upper half block character) and `<|im_end|>` has been replaced with `▄` (lower half block character).
49
 
 
44
  - [Context](https://huggingface.co/lemonilia/test-model-srp-0.5/blob/main/BlockML-Context.json)
45
  - [Instruct](https://huggingface.co/lemonilia/test-model-srp-0.5/resolve/main/BlockML-Instruct.json)
46
 
47
+ Note: the prompting format is **intentionally different** from that of the Mistral-Instruct base model.
48
+
49
  ## Block characters
50
  The model uses a _ChatML-like_ prompting format with a few changes from the usual roles typically used for ChatGPT-like assistant chatbots. The first is that to avoid complications with tokenization and special tokens, `<|im_start|>` has been replaced with `▀` (upper half block character) and `<|im_end|>` has been replaced with `▄` (lower half block character).
51