This is a merge of pre-trained language models created using
mergekit.
This model was merged using the Consensus-Orthogonal Residual Decomposition merge method using
Qwen/Qwen2.5-1.5B as a base.
1base_model: Qwen/Qwen2.5-1.5B
2dtype: bfloat16
3merge_method: cord
4modules:
5 default:
6 slices:
7 - sources:
8 - layer_range: [0, 28]
9 model: Qwen/Qwen2.5-1.5B-Instruct
10 parameters:
11 weight: 1.0
12 - layer_range: [0, 28]
13 model: Qwen/Qwen2.5-Coder-1.5B-Instruct
14 parameters:
15 weight: 1.0
16 - layer_range: [0, 28]
17 model: Qwen/Qwen2.5-1.5B
18parameters:
19 consensus_threshold: -1.0
20 lambda_consensus: 1.0
21 lambda_unique: -1.0
22 power_iters: 20.0
23 row_granularity: 1.0
24 sign_trim_consensus: 1.0
The function 'cord' method does not exist in MergeKit.
