Ba2han commited on
Commit
6531b97
1 Parent(s): 89b8403

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -16
README.md CHANGED
@@ -7,25 +7,11 @@ language:
7
  - en
8
  license: mit
9
  ---
10
- hf-causal-experimental (pretrained='E:/text-generation-webui/models/phi-984-dora'), limit: None, provide_description: False, num_fewshot: 5, batch_size: None
11
- | Task |Version| Metric |Value | |Stderr|
12
- |---------------------------------|------:|--------|-----:|---|-----:|
13
- |arc_challenge | 0|acc |0.5870|± |0.0144|
14
- | | |acc_norm|0.6067|± |0.0143|
15
- |hendrycksTest-abstract_algebra | 1|acc |0.3500|± |0.0479|
16
- | | |acc_norm|0.3500|± |0.0479|
17
- |hendrycksTest-college_biology | 1|acc |0.8264|± |0.0317|
18
- | | |acc_norm|0.8264|± |0.0317|
19
- |hendrycksTest-college_chemistry | 1|acc |**0.4900**|± |0.0502|
20
- | | |acc_norm|0.4900|± |0.0502|
21
- |hendrycksTest-college_mathematics| 1|acc |**0.3900**|± |0.0490|
22
- | | |acc_norm|0.3900|± |0.0490|
23
- |hendrycksTest-college_physics | 1|acc |**0.4020**|± |0.0488|
24
- | | |acc_norm|0.4020|± |0.0488|
25
- |winogrande | 0|acc |**0.7309**|± |0.0125|
26
 
27
  **We have Llama-3 at home!**
28
 
 
 
29
  The model has been trained on filtered versions of tagged datasets, as well as a few thousand more examples generated with llama-3-70B.
30
 
31
  Use **Zephyr template** with any system message. Default system message should be:
 
7
  - en
8
  license: mit
9
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
10
 
11
  **We have Llama-3 at home!**
12
 
13
+ Highest MMLU and Winogrande on the board!
14
+
15
  The model has been trained on filtered versions of tagged datasets, as well as a few thousand more examples generated with llama-3-70B.
16
 
17
  Use **Zephyr template** with any system message. Default system message should be: