Views
No views yet
1
2models:
3 - model: deepseek-ai/DeepSeek-R1-0528-Qwen3-8B
4 parameters:
5 weight: 1
6 - model: Qwen/Qwen3-8B
7 parameters:
8 weight: 0
9merge_method: linear
10base_model: deepseek-ai/DeepSeek-R1-0528-Qwen3-8B
11parameters:
12 normalize: false
13tokenizer:
14 source: "Qwen/Qwen3-8B"
15 tokens:
16 <|endoftext|>:
17 source: "Qwen/Qwen3-8B"
18 <|im_end|>:
19 source: "Qwen/Qwen3-8B"
20 <|end▁of▁sentence|>:
21 source:
22 model: "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B"
23 kind: "model_token"
24 token: "<|end▁of▁sentence|>"
25 force: true
26 <|User|>:
27 source:
28 model: "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B"
29 kind: "model_token"
30 token: "<|User|>"
31 force: true
32 <|Assistant|>:
33 source:
34 model: "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B"
35 kind: "model_token"
36 token: "<|Assistant|>"
37 force: true
38 <|begin▁of▁sentence|>:
39 source:
40 model: "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B"
41 kind: "model_token"
42 token: "<|begin▁of▁sentence|>"
43 force: true
44dtype: bfloat16
45
46