File size: 1,476 Bytes
4f91585
 
 
 
 
 
 
 
 
 
 
 
 
 
 
dcf52e6
 
 
 
 
f140efc
 
7a262c0
 
4f91585
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
---
base_model:
- Qwen/Qwen2.5-14B-Instruct
- Lambent/alternate-instruct-qwen2.5-14B
library_name: transformers
tags:
- mergekit
- merge

---
# qwenselfinstructalt

This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).

## Merge Details

Same idea as Lambent/qwen2.5-14B-selfmerge-A, but training the base model on an ~20M token instruct and continued pretraining dataset first.

Hope is the lightweight instruction tuning might add some synergy with the original instruct.

Testing: eq-bench showed no syntax errors and result was 75.6984, closer to original instruct value of 76.9195 than selfmerge-A (which had 73.8068).

Subsets of mrfakename/Capybara-ShareGPT, abacusai/SystemChat-1.1, anthracite-org/nopm_claude_writing_fixed and fineweb-edu were used for the alternate training.

### Merge Method

This model was merged using the SLERP merge method.

### Models Merged

The following models were included in the merge:
* [Qwen/Qwen2.5-14B-Instruct](https://huggingface.co/Qwen/Qwen2.5-14B-Instruct)
* [Lambent/alternate-instruct-qwen2.5-14B](https://huggingface.co/Lambent/alternate-instruct-qwen2.5-14B)

### Configuration

The following YAML configuration was used to produce this model:

```yaml
models:
  - model: Lambent/alternate-instruct-qwen2.5-14B
merge_method: slerp
base_model: Qwen/Qwen2.5-14B-Instruct
parameters:
  t:
    - value: [0, 0, 0.3, 0.4, 0.5, 0.6, 0.5, 0.4, 0.3, 0, 0]
dtype: bfloat16


```