© 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. 35-TRANSCODER-HP
    7. 39326
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    Not in Any Lists

    No Comments

    Negative Logits
    sie
    -0.28
    ifter
    -0.28
    oire
    -0.28
    afort
    -0.28
    ç«ĭæĸ¹
    -0.28
    vided
    -0.27
    aise
    -0.27
    icer
    -0.26
    apas
    -0.26
    gesch
    -0.26
    POSITIVE LOGITS
     end
    0.34
     already
    0.33
     pre
    0.33
     pres
    0.31
     un
    0.31
     a
    0.31
     extra
    0.30
     won
    0.30
     non
    0.30
     whole
    0.30
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.