gghfez/GLM-4.6-control-vectors

Creative Writing control-vectors for zai-org/GLM-4.6

Usage example

llama.cpp / ik_llama.cpp

Creative writing:

llama-server --model  GLM-4.6-UD-IQ2_XXS-00001-of-00003.gguf [other CLI arguments] \
    --control-vector-scaled  GLM-4.6-honesty_vs_machiavellianism__debias.gguf 1.0 \
    --control-vector-scaled  GLM-4.6-honesty_vs_machiavellianism__machiavellianism.gguf 1.0 \

Assistant:

llama-server --model  GLM-4.6-IQ3_KS-00001-of-00004.gguf [other CLI arguments] \
    --control-vector-scaled  GLM-4.6-communication__debias.gguf 1.0 \
    --control-vector-scaled  GLM-4.6-communication__direct_communication.gguf 1.0 \

TODO: upload after I get below the huggingface storage limit.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for gghfez/GLM-4.6-control-vectors

Base model

zai-org/GLM-4.6
Finetuned
(5)
this model