Views
No views yet
1slices:
2 - sources:
3 - model: Qwen/Qwen2.5-3B-Instruct
4 layer_range: [0, 4]
5 - sources:
6 - model: Qwen/Qwen2.5-3B-Instruct
7 layer_range: [2, 6]
8 - sources:
9 - model: Qwen/Qwen2.5-3B-Instruct
10 layer_range: [4, 8]
11 - sources:
12 - model: Qwen/Qwen2.5-3B-Instruct
13 layer_range: [6, 10]
14 - sources:
15 - model: Qwen/Qwen2.5-3B-Instruct
16 layer_range: [8, 12]
17 - sources:
18 - model: Qwen/Qwen2.5-3B-Instruct
19 layer_range: [10, 14]
20 - sources:
21 - model: Qwen/Qwen2.5-3B-Instruct
22 layer_range: [12, 16]
23 - sources:
24 - model: Qwen/Qwen2.5-3B-Instruct
25 layer_range: [14, 18]
26 - sources:
27 - model: Qwen/Qwen2.5-3B-Instruct
28 layer_range: [16, 20]
29 - sources:
30 - model: Qwen/Qwen2.5-3B-Instruct
31 layer_range: [18, 22]
32 - sources:
33 - model: Qwen/Qwen2.5-3B-Instruct
34 layer_range: [20, 24]
35 - sources:
36 - model: Qwen/Qwen2.5-3B-Instruct
37 layer_range: [22, 26]
38 - sources:
39 - model: Qwen/Qwen2.5-3B-Instruct
40 layer_range: [24, 28]
41 - sources:
42 - model: Qwen/Qwen2.5-3B-Instruct
43 layer_range: [26, 30]
44 - sources:
45 - model: Qwen/Qwen2.5-3B-Instruct
46 layer_range: [28, 32]
47 - sources:
48 - model: Qwen/Qwen2.5-3B-Instruct
49 layer_range: [30, 34]
50 - sources:
51 - model: Qwen/Qwen2.5-3B-Instruct
52 layer_range: [32, 36]
53tokenizer_source: Qwen/Qwen2.5-3B-Instruct
54merge_method: passthrough
55dtype: bfloat16