Hugging Face's logo

nm-testing
/
Llama-2-7b-hf-pruned50-quant-ds

Text Generation
Transformers
ONNX
llama
deepsparse
Model card Files Community
mwitiderrick commited on
Commit
58516fe
•
1 Parent(s): dd4f40e

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -6
README.md CHANGED
@@ -3,9 +3,9 @@ base_model: NousResearch/Llama-2-7b-hf
3
  inference: false
4
  model_type: llama
5
  prompt_template: |
6
- ### Instruction:
7
  {prompt}
8
- ### Response:
9
  quantized_by: mwitiderrick
10
  tags:
11
  - deepsparse
@@ -37,11 +37,9 @@ print(model(formatted_prompt, max_new_tokens=200).generations[0].text)
37
 
38
  ## Prompt template
39
  ```
40
- ### Instruction:
41
  <prompt>
42
-
43
- ### Response:
44
- <leave a newline blank for model to respond>
45
 
46
  ```
47
  ## Sparsification
 
3
  inference: false
4
  model_type: llama
5
  prompt_template: |
6
+ <s>[INST]
7
  {prompt}
8
+ [/INST]
9
  quantized_by: mwitiderrick
10
  tags:
11
  - deepsparse
 
37
 
38
  ## Prompt template
39
  ```
40
+ <s>[INST]
41
  <prompt>
42
+ [/INST]
 
 
43
 
44
  ```
45
  ## Sparsification