cookinai commited on
Commit
1bca790
1 Parent(s): 7d39449

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +26 -0
README.md CHANGED
@@ -1,3 +1,29 @@
1
  ---
2
  license: apache-2.0
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: apache-2.0
3
  ---
4
+ Slerp Merge of shadowml/Marcoro14-7B-slerp and rishiraj/CatPPT
5
+ I've been meaning to mix in EmbeddedLLM/Mistral-7B-Merge-14-v0.1 but have had issues so thanks to shadowml that merges it with AIDC-ai-business/Marcoroni-7B-v3
6
+
7
+ Also, been hearing talks of AIDC-ai-business/Marcoroni-7B-v3 being contaminated,
8
+ I don't know if this is true but make a post on HuggingFaceH4/open_llm_leaderboard so we can keep the board clean
9
+
10
+ .yaml file for mergekit
11
+
12
+ ```.yaml:
13
+ slices:
14
+ - sources:
15
+ - model: shadowml/Marcoro14-7B-slerp
16
+ layer_range: [0, 32]
17
+ - model: rishiraj/CatPPT
18
+ layer_range: [0, 32]
19
+ merge_method: slerp
20
+ base_model: shadowml/Marcoro14-7B-slerp
21
+ parameters:
22
+ t:
23
+ - filter: self_attn
24
+ value: [0, 0.5, 0.3, 0.7, 1]
25
+ - filter: mlp
26
+ value: [1, 0.5, 0.7, 0.3, 0]
27
+ - value: 0.5 # fallback for rest of tensors
28
+ dtype: bfloat16
29
+ ```