v000000 commited on
Commit
1c29bfa
·
verified ·
1 Parent(s): 3a89443

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +37 -0
README.md CHANGED
@@ -32,6 +32,43 @@ The following models were included in the merge:
32
  The following YAML configuration was used to produce this model:
33
 
34
  ```yaml
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
35
  models:
36
  - model: v000000/L3.1-8B-RP-Test-003-Task_Arithmetic
37
  merge_method: slerp
 
32
  The following YAML configuration was used to produce this model:
33
 
34
  ```yaml
35
+ slices:
36
+ - sources:
37
+ - model: Sao10K/L3.1-8B-Niitama-v1.1+grimjim/Llama-3-Instruct-abliteration-LoRA-8B
38
+ layer_range: [0, 32]
39
+ - model: akjindal53244/Llama-3.1-Storm-8B
40
+ layer_range: [0, 32]
41
+ merge_method: nearswap
42
+ base_model: Sao10K/L3.1-8B-Niitama-v1.1+grimjim/Llama-3-Instruct-abliteration-LoRA-8B
43
+ parameters:
44
+ t:
45
+ - value: 0.0001
46
+ dtype: bfloat16
47
+ out_type: float16 #oops
48
+ models:
49
+ - model: arcee-ai/Llama-3.1-SuperNova-Lite
50
+ parameters:
51
+ weight: 1.0
52
+ - model: v000000/L3.1-Niitorm-8B-t0.0001
53
+ parameters:
54
+ weight: 0.4
55
+ merge_method: task_arithmetic
56
+ base_model: arcee-ai/Llama-3.1-SuperNova-Lite
57
+ parameters:
58
+ normalize: false
59
+ dtype: float16
60
+ models:
61
+ - model: arcee-ai/Llama-3.1-SuperNova-Lite
62
+ parameters:
63
+ weight: 0.0
64
+ - model: v000000/L3.1-Niitorm-8B-t0.0001
65
+ parameters:
66
+ weight: 1.25
67
+ merge_method: task_arithmetic
68
+ base_model: arcee-ai/Llama-3.1-SuperNova-Lite
69
+ parameters:
70
+ normalize: false
71
+ dtype: float16
72
  models:
73
  - model: v000000/L3.1-8B-RP-Test-003-Task_Arithmetic
74
  merge_method: slerp