de-llama3-discoleo-instruct-8b-v0.1-x-meta-llama-3-8b-instruct-ffn_kv_injection
/
mergekit_config.yml
| dtype: bfloat16 | |
| tokenizer: | |
| source: union | |
| merge_method: ffn_kv_injection | |
| base_model: meta-llama/Meta-Llama-3-8B-Instruct | |
| models: | |
| - model: meta-llama/Meta-Llama-3-8B-Instruct | |
| parameters: {} | |
| - model: DiscoResearch/Llama3-DiscoLeo-Instruct-8B-v0.1 | |
| parameters: {} | |
| parameters: | |
| weights: | |
| - 0.7 | |
| - 0.3 | |
| tau: 1.0 | |
| aspect_thresh: 1.5 | |
| pnorm: 2.0 | |
| alpha_floor: 0.0 | |
| alpha_ceil: 1.0 | |
| write_readme: README.md | |