© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 32-TRANSCODER-HP
    7. 41455
    Prev
    Next
    INDEX
    Explanations

    medication/chemistry

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    åħĭæĭī
    -0.29
    hor
    -0.28
    ipel
    -0.24
     Há»ĵ
    -0.24
    akedirs
    -0.24
     Reco
    -0.24
    AL
    -0.24
     Horm
    -0.24
    赫
    -0.24
     Hidden
    -0.23
    POSITIVE LOGITS
     cy
    0.30
     DROP
    0.29
    utivo
    0.28
    追éĢIJ
    0.28
    Cy
    0.26
    cy
    0.26
    第ä¸Ģ天
    0.26
    ucker
    0.26
     Dro
    0.25
    伪
    0.25
    Activations Density 0.004%

    No Known Activations