© 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. 18099
    Prev
    Next
    INDEX
    Explanations

    ost

    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
    roma
    -0.30
    anced
    -0.29
    App
    -0.27
    å¸ļ
    -0.26
    zin
    -0.26
    rho
    -0.26
     ros
    -0.25
    çĦ¯
    -0.25
    寤
    -0.25
    UIColor
    -0.24
    POSITIVE LOGITS
    hen
    0.31
    elles
    0.29
    alles
    0.28
    ï¿ł
    0.27
    olle
    0.27
    yne
    0.26
    elle
    0.26
    allee
    0.25
    REFERRED
    0.25
    imson
    0.25
    Activations Density 0.011%

    No Known Activations