Doesnt work with llama.cpp convert_hf_to_gguf.py yet, but should when support comes
Refusals: 3/100, KL divergence: 0.0407
- direction_index = 16.33
- attn.o_proj.max_weight = 1.39
- attn.o_proj.max_weight_position = 16.52
- attn.o_proj.min_weight = 1.25
- attn.o_proj.min_weight_distance = 16.04
- mlp.down_proj.max_weight = 1.14
- mlp.down_proj.max_weight_position = 16.53
- mlp.down_proj.min_weight = 0.68
- mlp.down_proj.min_weight_distance = 10.29
- Downloads last month
- -
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for CCATresearch/North-Micro-Vision-Instruct-HERETIC
Base model
CohereLabs/North-Micro-Vision-Instruct