© 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. 18051
    Prev
    Next
    INDEX
    Explanations

    is/be variations

    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
    иÑĢовки
    -0.28
    ableView
    -0.28
    pute
    -0.28
    adium
    -0.27
    é»Ķ
    -0.26
    gger
    -0.26
     Guinness
    -0.26
    ABL
    -0.25
    MX
    -0.24
     gros
    -0.24
    POSITIVE LOGITS
    æĮij
    0.41
     pick
    0.35
     picking
    0.35
     picked
    0.33
     dispersed
    0.32
     picks
    0.31
     Pick
    0.31
    _pick
    0.30
     occurring
    0.29
     responsible
    0.29
    Activations Density 0.297%

    No Known Activations