Suparious commited on
Commit
a6f60ea
·
verified ·
1 Parent(s): 9a9dc1b

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +29 -2
README.md CHANGED
@@ -1,4 +1,5 @@
1
  ---
 
2
  library_name: transformers
3
  tags:
4
  - 4-bit
@@ -10,6 +11,32 @@ pipeline_tag: text-generation
10
  inference: false
11
  quantized_by: Suparious
12
  ---
13
- #
14
 
15
- **UPLOAD IN PROGRESS**
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ license: apache-2.0
3
  library_name: transformers
4
  tags:
5
  - 4-bit
 
11
  inference: false
12
  quantized_by: Suparious
13
  ---
14
+ # dawn17/MistarlingMaid-2x7B-base AWQ
15
 
16
+ - Model creator: [dawn17](https://huggingface.co/dawn17)
17
+ - Original model: [MistarlingMaid-2x7B-base](https://huggingface.co/dawn17/MistarlingMaid-2x7B-base)
18
+
19
+ ## Model Summary
20
+
21
+ base_model: /Users/dawn/git/models/Mistral-7B-Instruct-v0.2
22
+ gate_mode: hidden # one of "hidden", "cheap_embed", or "random"
23
+ dtype: bfloat16 # output dtype (float32, float16, or bfloat16)
24
+ experts:
25
+ - source_model: /Users/dawn/git/models/Silicon-Maid-7B
26
+ positive_prompts:
27
+ - "roleplay"
28
+ - source_model: /Users/dawn/git/models/Starling-LM-7B-beta
29
+ positive_prompts:
30
+ - "chat"
31
+
32
+ - [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
33
+
34
+ | Metric |Value|
35
+ |---------------------------------|----:|
36
+ |Avg. |68.01|
37
+ |AI2 Reasoning Challenge (25-Shot)|67.49|
38
+ |HellaSwag (10-Shot) |84.76|
39
+ |MMLU (5-Shot) |62.62|
40
+ |TruthfulQA (0-shot) |58.93|
41
+ |Winogrande (5-shot) |78.22|
42
+ |GSM8k (5-shot) |56.03|