Capability-head weights used in the Multi-Head Latent Control paper. Paper: https://arxiv.org/abs/2607.14277
Amirhosein Ghasemabadi
AmirhoseinGH
AI & ML interests
None yet
Recent Activity
upvoted a paper 2 days ago
Multi-Head Latent Control: A Unified Interface for LLM Agent Decision Making submitted a paper 2 days ago
Multi-Head Latent Control: A Unified Interface for LLM Agent Decision Making updated a collection 5 days ago
Multi Head Latent Control training dataOrganizations
Training data release for Multi Head Latent Control. Paper: https://arxiv.org/abs/2607.14277
-
AmirhoseinGH/mhlc-training-gemma4-gemma4_e4b_it_think_off_hard_mixed_sources_120k
Viewer • Updated • 116k • 70 -
AmirhoseinGH/mhlc-training-gemma4-gemma4_e4b_it_think_on_hard_mixed_sources_120k
Viewer • Updated • 116k • 56 -
AmirhoseinGH/mhlc-training-qwen3.5-qwen3_5_4b_think_off_hard_mixed_sources_120k
Viewer • Updated • 116k • 74 -
AmirhoseinGH/mhlc-training-qwen3.5-qwen3_5_9b_think_off_hard_mixed_sources_120k
Viewer • Updated • 116k • 86
Capability-head weights used in the Multi-Head Latent Control paper. Paper: https://arxiv.org/abs/2607.14277
Training data release for Multi Head Latent Control. Paper: https://arxiv.org/abs/2607.14277
-
AmirhoseinGH/mhlc-training-gemma4-gemma4_e4b_it_think_off_hard_mixed_sources_120k
Viewer • Updated • 116k • 70 -
AmirhoseinGH/mhlc-training-gemma4-gemma4_e4b_it_think_on_hard_mixed_sources_120k
Viewer • Updated • 116k • 56 -
AmirhoseinGH/mhlc-training-qwen3.5-qwen3_5_4b_think_off_hard_mixed_sources_120k
Viewer • Updated • 116k • 74 -
AmirhoseinGH/mhlc-training-qwen3.5-qwen3_5_9b_think_off_hard_mixed_sources_120k
Viewer • Updated • 116k • 86
models 18
AmirhoseinGH/mhlc-capability-head-gemma4-e4b-thinking-prefix200
Updated
AmirhoseinGH/mhlc-capability-head-qwen3vl-4b-thinking-prefix200
Updated
AmirhoseinGH/mhlc-capability-head-qwen3vl-2b-thinking-prefix200
Updated
AmirhoseinGH/mhlc-capability-head-qwen3vl-4b-thinking-lite
Updated
AmirhoseinGH/mhlc-capability-head-qwen3vl-32b-instruct-step10000
Updated
AmirhoseinGH/mhlc-capability-head-gemma4-e4b-thinking
Updated
AmirhoseinGH/mhlc-capability-head-gemma4-e4b-instruct
Updated
AmirhoseinGH/mhlc-capability-head-qwen35-9b
Updated
AmirhoseinGH/mhlc-capability-head-qwen35-4b
Updated
AmirhoseinGH/mhlc-capability-head-qwen3vl-4b-thinking
Updated
datasets 18
AmirhoseinGH/mhlc-training-qwen3vl-qwen3_vl_4b_thinking_hard_mixed_sources_120k
Viewer • Updated • 116k • 71
AmirhoseinGH/mhlc-training-qwen3vl-qwen3_vl_4b_instruct_hard_mixed_sources_120k
Viewer • Updated • 116k • 94
AmirhoseinGH/mhlc-training-qwen3vl-qwen3_vl_2b_thinking_hard_mixed_sources_120k
Viewer • Updated • 115k • 71
AmirhoseinGH/mhlc-training-qwen3vl-qwen3_vl_2b_instruct_hard_mixed_sources_120k
Viewer • Updated • 115k • 47
AmirhoseinGH/mhlc-training-qwen3vl-qwen3-vl-32b-instruct_hard_mixed_sources_120k
Viewer • Updated • 84k • 70
AmirhoseinGH/mhlc-training-qwen3.5-qwen3_5_9b_think_off_hard_mixed_sources_120k
Viewer • Updated • 116k • 86
AmirhoseinGH/mhlc-training-qwen3.5-qwen3_5_4b_think_off_hard_mixed_sources_120k
Viewer • Updated • 116k • 74
AmirhoseinGH/mhlc-training-gemma4-gemma4_e4b_it_think_on_hard_mixed_sources_120k
Viewer • Updated • 116k • 56
AmirhoseinGH/mhlc-training-gemma4-gemma4_e4b_it_think_off_hard_mixed_sources_120k
Viewer • Updated • 116k • 70
AmirhoseinGH/when2call-qwen3vl-2b-multiclass
Viewer • Updated • 32k • 21