pszemraj commited on
Commit
66da35c
·
verified ·
1 Parent(s): ab87b06

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +14 -14
README.md CHANGED
@@ -17,26 +17,26 @@ base_model:
17
 
18
  | Parameter | Value |
19
  | :-------- | :---: |
20
- | **direction_index** | 24.65 |
21
- | **attn.o_proj.max_weight** | 1.13 |
22
- | **attn.o_proj.max_weight_position** | 30.77 |
23
- | **attn.o_proj.min_weight** | 0.50 |
24
- | **attn.o_proj.min_weight_distance** | 17.16 |
25
- | **mamba.out_proj.max_weight** | 1.44 |
26
- | **mamba.out_proj.max_weight_position** | 26.90 |
27
- | **mamba.out_proj.min_weight** | 0.91 |
28
- | **mamba.out_proj.min_weight_distance** | 19.24 |
29
- | **mlp.shared_down_proj.max_weight** | 0.86 |
30
- | **mlp.shared_down_proj.max_weight_position** | 28.04 |
31
- | **mlp.shared_down_proj.min_weight** | 0.00 |
32
- | **mlp.shared_down_proj.min_weight_distance** | 13.42 |
33
 
34
  ## Performance
35
 
36
  | Metric | This model | Original model ([ibm-granite/granite-4.0-h-1b](https://huggingface.co/ibm-granite/granite-4.0-h-1b)) |
37
  | :----- | :--------: | :---------------------------: |
38
  | **KL divergence** | 0.03 | 0 *(by definition)* |
39
- | **Refusals** | 12/100 | 93/100 |
40
 
41
  -----
42
 
 
17
 
18
  | Parameter | Value |
19
  | :-------- | :---: |
20
+ | **direction_index** | 23.75 |
21
+ | **attn.o_proj.max_weight** | 1.75 |
22
+ | **attn.o_proj.max_weight_position** | 24.08 |
23
+ | **attn.o_proj.min_weight** | 1.17 |
24
+ | **attn.o_proj.min_weight_distance** | 20.46 |
25
+ | **mamba.out_proj.max_weight** | 1.86 |
26
+ | **mamba.out_proj.max_weight_position** | 31.62 |
27
+ | **mamba.out_proj.min_weight** | 1.29 |
28
+ | **mamba.out_proj.min_weight_distance** | 18.93 |
29
+ | **mlp.shared_down_proj.max_weight** | 1.11 |
30
+ | **mlp.shared_down_proj.max_weight_position** | 36.36 |
31
+ | **mlp.shared_down_proj.min_weight** | 0.16 |
32
+ | **mlp.shared_down_proj.min_weight_distance** | 4.17 |
33
 
34
  ## Performance
35
 
36
  | Metric | This model | Original model ([ibm-granite/granite-4.0-h-1b](https://huggingface.co/ibm-granite/granite-4.0-h-1b)) |
37
  | :----- | :--------: | :---------------------------: |
38
  | **KL divergence** | 0.03 | 0 *(by definition)* |
39
+ | **Refusals** | 7/100 | 93/100 |
40
 
41
  -----
42