| models: | |
| - model: deepseek-ai/DeepSeek-R1-Distill-Llama-70B | |
| parameters: | |
| weight: 0.25 | |
| density: 0.5 | |
| - model: watt-ai/watt-tool-70B | |
| parameters: | |
| weight: 0.25 | |
| density: 0.5 | |
| - model: deepcogito/cogito-v1-preview-llama-70B | |
| parameters: | |
| weight: 0.25 | |
| density: 0.5 | |
| - model: nvidia/Llama-3.1-Nemotron-70B-Instruct-HF | |
| parameters: | |
| weight: 0.25 | |
| density: 0.5 | |
| merge_method: dare_ties | |
| base_model: meta-llama/Llama-3.3-70B-Instruct | |
| parameters: | |
| normalize: false | |
| out_dtype: bfloat16 | |
| chat_template: llama3 | |
| tokenizer: | |
| source: base | |
| pad_to_multiple_of: 8 |