Views
No views yet
1# Part 3, Apply abliteration (SwallowMaid-8B)
2models:
3 - model: sppo-rpmix-part2+grimjim/Llama-3-Instruct-abliteration-LoRA-8B
4 parameters:
5 weight: 1.0
6merge_method: linear
7dtype: float32
8
9# Part 2, infuse 35% swallow+rpmix to SPPO-Iter3 (sppo-rpmix-part2)
10models:
11 - model: UCLA-AGI/Llama-3-Instruct-8B-SPPO-Iter3
12 parameters:
13 weight: 1.0
14 - model: rpmix-part1
15 parameters:
16 weight: 0.35
17merge_method: task_arithmetic
18base_model: UCLA-AGI/Llama-3-Instruct-8B-SPPO-Iter3
19parameters:
20 normalize: false
21dtype: float32
22
23# Part 1, linear merge rpmix (rpmix-part1)
24models:
25 - model: Nitral-AI/Hathor_Respawn-L3-8B-v0.8
26 parameters:
27 weight: 0.6
28 - model: maldv/llama-3-fantasy-writer-8b
29 parameters:
30 weight: 0.1
31 - model: NeverSleep/Llama-3-Lumimaid-8B-v0.1-OAS
32 parameters:
33 weight: 0.4
34 - model: tokyotech-llm/Llama-3-Swallow-8B-v0.1
35 parameters:
36 weight: 0.15
37merge_method: linear
38dtype: float321<|begin_of_text|><|start_header_id|>system<|end_header_id|>
2
3{system_prompt}<|eot_id|><|start_header_id|>user<|end_header_id|>
4
5{input}<|eot_id|><|start_header_id|>assistant<|end_header_id|>
6
7{output}<|eot_id|>
8