© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 35-TRANSCODER-HP
    7. 38529
    Prev
    Next
    INDEX
    Explanations

    Her

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    é¼ĵèĪŀ
    -0.28
    å¸Ĥåľºç«ŀäºī
    -0.28
    åİ»
    -0.26
    ,'\
    -0.24
    以ä¸ĭåĩłä¸ª
    -0.23
    大æī¹
    -0.23
    èĢĮåİ»
    -0.23
     unsub
    -0.23
    åİ»çľĭçľĭ
    -0.23
    åĮºåŁŁæĢ§
    -0.23
    POSITIVE LOGITS
    d
    0.36
    dom
    0.36
    dh
    0.36
    dy
    0.36
    dv
    0.36
    do
    0.36
    dto
    0.35
    days
    0.35
    db
    0.33
    dt
    0.33
    Activations Density 3.114%

    No Known Activations