© 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. 31-TRANSCODER-HP
    7. 19647
    Prev
    Next
    INDEX
    Explanations

    Non-standard text

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    çĽĸ
    -0.33
    èĵĭ
    -0.31
    Serv
    -0.28
    bservice
    -0.28
     thunk
    -0.28
    onte
    -0.27
    ằ
    -0.27
     çĻ»
    -0.27
    peare
    -0.26
    oce
    -0.26
    POSITIVE LOGITS
    åĴ¨
    0.28
    -spin
    0.26
    èħIJèļĢ
    0.26
    è¿ĻæĿ¡
    0.26
    anio
    0.26
    å¾ģæĶ¶
    0.25
    两æĿ¡
    0.24
    upplier
    0.24
    _chain
    0.24
    (ll
    0.24
    Activations Density 1.081%

    No Known Activations