Files changed (1) hide show
  1. README.md +87 -74
README.md CHANGED
@@ -1,74 +1,87 @@
1
- ---
2
- base_model:
3
- - Qwen/Qwen2.5-3B-Instruct
4
- - ehristoforu/flc-r-0004-lora
5
- - Qwen/Qwen2.5-3B
6
- - Qwen/Qwen2.5-3B-Instruct
7
- - Qwen/Qwen2.5-3B-Instruct
8
- - ehristoforu/flc-r-0001-lora
9
- - Qwen/Qwen2.5-3B-Instruct
10
- - ehristoforu/flc-r-0002-lora
11
- - Qwen/Qwen2.5-3B-Instruct
12
- - ehristoforu/flc-r-0003-lora
13
- library_name: transformers
14
- tags:
15
- - mergekit
16
- - merge
17
-
18
- ---
19
- # merge
20
-
21
- This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
22
-
23
- ## Merge Details
24
- ### Merge Method
25
-
26
- This model was merged using the [TIES](https://arxiv.org/abs/2306.01708) merge method using [Qwen/Qwen2.5-3B](https://huggingface.co/Qwen/Qwen2.5-3B) as a base.
27
-
28
- ### Models Merged
29
-
30
- The following models were included in the merge:
31
- * [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct) + [ehristoforu/flc-r-0004-lora](https://huggingface.co/ehristoforu/flc-r-0004-lora)
32
- * [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct)
33
- * [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct) + [ehristoforu/flc-r-0001-lora](https://huggingface.co/ehristoforu/flc-r-0001-lora)
34
- * [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct) + [ehristoforu/flc-r-0002-lora](https://huggingface.co/ehristoforu/flc-r-0002-lora)
35
- * [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct) + [ehristoforu/flc-r-0003-lora](https://huggingface.co/ehristoforu/flc-r-0003-lora)
36
-
37
- ### Configuration
38
-
39
- The following YAML configuration was used to produce this model:
40
-
41
- ```yaml
42
- models:
43
- - model: Qwen/Qwen2.5-3B-Instruct+ehristoforu/flc-r-0001-lora
44
- parameters:
45
- weight: 1
46
- density: 1
47
- - model: Qwen/Qwen2.5-3B-Instruct+ehristoforu/flc-r-0002-lora
48
- parameters:
49
- weight: 1
50
- density: 1
51
- - model: Qwen/Qwen2.5-3B-Instruct+ehristoforu/flc-r-0003-lora
52
- parameters:
53
- weight: 1
54
- density: 1
55
- - model: Qwen/Qwen2.5-3B-Instruct+ehristoforu/flc-r-0004-lora
56
- parameters:
57
- weight: 1
58
- density: 1
59
- - model: Qwen/Qwen2.5-3B-Instruct
60
- parameters:
61
- weight: 1
62
- density: 1
63
- merge_method: ties
64
- base_model: Qwen/Qwen2.5-3B
65
- parameters:
66
- weight: 1
67
- density: 1
68
- normalize: true
69
- int8_mask: true
70
- tokenizer_source: Qwen/Qwen2.5-3B-Instruct
71
- dtype: float16
72
-
73
-
74
- ```
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model:
3
+ - Qwen/Qwen2.5-3B-Instruct
4
+ - ehristoforu/flc-r-0004-lora
5
+ - Qwen/Qwen2.5-3B
6
+ - Qwen/Qwen2.5-3B-Instruct
7
+ - Qwen/Qwen2.5-3B-Instruct
8
+ - ehristoforu/flc-r-0001-lora
9
+ - Qwen/Qwen2.5-3B-Instruct
10
+ - ehristoforu/flc-r-0002-lora
11
+ - Qwen/Qwen2.5-3B-Instruct
12
+ - ehristoforu/flc-r-0003-lora
13
+ library_name: transformers
14
+ tags:
15
+ - mergekit
16
+ - merge
17
+ language:
18
+ - zho
19
+ - eng
20
+ - fra
21
+ - spa
22
+ - por
23
+ - deu
24
+ - ita
25
+ - rus
26
+ - jpn
27
+ - kor
28
+ - vie
29
+ - tha
30
+ - ara
31
+ ---
32
+ # merge
33
+
34
+ This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
35
+
36
+ ## Merge Details
37
+ ### Merge Method
38
+
39
+ This model was merged using the [TIES](https://arxiv.org/abs/2306.01708) merge method using [Qwen/Qwen2.5-3B](https://huggingface.co/Qwen/Qwen2.5-3B) as a base.
40
+
41
+ ### Models Merged
42
+
43
+ The following models were included in the merge:
44
+ * [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct) + [ehristoforu/flc-r-0004-lora](https://huggingface.co/ehristoforu/flc-r-0004-lora)
45
+ * [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct)
46
+ * [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct) + [ehristoforu/flc-r-0001-lora](https://huggingface.co/ehristoforu/flc-r-0001-lora)
47
+ * [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct) + [ehristoforu/flc-r-0002-lora](https://huggingface.co/ehristoforu/flc-r-0002-lora)
48
+ * [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct) + [ehristoforu/flc-r-0003-lora](https://huggingface.co/ehristoforu/flc-r-0003-lora)
49
+
50
+ ### Configuration
51
+
52
+ The following YAML configuration was used to produce this model:
53
+
54
+ ```yaml
55
+ models:
56
+ - model: Qwen/Qwen2.5-3B-Instruct+ehristoforu/flc-r-0001-lora
57
+ parameters:
58
+ weight: 1
59
+ density: 1
60
+ - model: Qwen/Qwen2.5-3B-Instruct+ehristoforu/flc-r-0002-lora
61
+ parameters:
62
+ weight: 1
63
+ density: 1
64
+ - model: Qwen/Qwen2.5-3B-Instruct+ehristoforu/flc-r-0003-lora
65
+ parameters:
66
+ weight: 1
67
+ density: 1
68
+ - model: Qwen/Qwen2.5-3B-Instruct+ehristoforu/flc-r-0004-lora
69
+ parameters:
70
+ weight: 1
71
+ density: 1
72
+ - model: Qwen/Qwen2.5-3B-Instruct
73
+ parameters:
74
+ weight: 1
75
+ density: 1
76
+ merge_method: ties
77
+ base_model: Qwen/Qwen2.5-3B
78
+ parameters:
79
+ weight: 1
80
+ density: 1
81
+ normalize: true
82
+ int8_mask: true
83
+ tokenizer_source: Qwen/Qwen2.5-3B-Instruct
84
+ dtype: float16
85
+
86
+
87
+ ```