GGUF
jan-hq commited on
Commit
27cb8d1
·
verified ·
1 Parent(s): f9a6ce5

Create model.yml

Browse files
Files changed (1) hide show
  1. model.yml +20 -0
model.yml ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ name: mixtral
2
+ model: mixtral:7x8B
3
+ version: 1
4
+
5
+ files:
6
+ - llama_model_path: model.gguf
7
+
8
+ # Results Preferences
9
+ top_p: 0.95
10
+ temperature: 0.7
11
+ frequency_penalty: 0
12
+ presence_penalty: 0
13
+ max_tokens: 32768 # Infer from base config.json -> max_position_embeddings
14
+ stream: true # true | false
15
+
16
+ # Engine / Model Settings
17
+ ngl: 33 # Infer from base config.json -> num_attention_heads
18
+ ctx_len: 32768 # Infer from base config.json -> max_position_embeddings
19
+ engine: cortex.llamacpp
20
+ prompt_template: "[INST] {prompt} [/INST]"