TheBloke commited on
Commit
d207702
·
1 Parent(s): efb8dbc

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +8 -1
README.md CHANGED
@@ -34,6 +34,13 @@ GGML files are for CPU + GPU inference using [llama.cpp](https://github.com/gger
34
  * [2, 3, 4, 5, 6 and 8-bit GGML models for CPU+GPU inference](https://huggingface.co/TheBloke/baichuan-vicuna-7B-GGML)
35
  * [Unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/fireballoon/baichuan-vicuna-7b)
36
 
 
 
 
 
 
 
 
37
  <!-- compatibility_ggml start -->
38
  ## Compatibility
39
 
@@ -87,7 +94,7 @@ Refer to the Provided Files table below to see what files use which methods, and
87
  I use the following command line; adjust for your tastes and needs:
88
 
89
  ```
90
- ./main -t 10 -ngl 32 -m baichuan-vicuna-7b.ggmlv3.q5_0.bin --color -c 2048 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "### Instruction: Write a story about llamas\n### Response:"
91
  ```
92
  If you're able to use full GPU offloading, you should use `-t 1` to get best performance.
93
 
 
34
  * [2, 3, 4, 5, 6 and 8-bit GGML models for CPU+GPU inference](https://huggingface.co/TheBloke/baichuan-vicuna-7B-GGML)
35
  * [Unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/fireballoon/baichuan-vicuna-7b)
36
 
37
+ ## Prompt template: Vicuna 1.1
38
+ ```
39
+ A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.
40
+ USER: prompt
41
+ ASSISTANT:
42
+ ```
43
+
44
  <!-- compatibility_ggml start -->
45
  ## Compatibility
46
 
 
94
  I use the following command line; adjust for your tastes and needs:
95
 
96
  ```
97
+ ./main -t 10 -ngl 32 -m baichuan-vicuna-7b.ggmlv3.q5_0.bin --color -c 2048 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "USER: Write a story about llamas\nASSISTANT:"
98
  ```
99
  If you're able to use full GPU offloading, you should use `-t 1` to get best performance.
100