© 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. 33-TRANSCODER-HP
    7. 54333
    Prev
    Next
    INDEX
    Explanations

    locations

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    еÑĢж
    -0.27
    -cols
    -0.26
    åĽ½ç±į
    -0.25
    igans
    -0.25
    AAF
    -0.25
    enci
    -0.25
    ament
    -0.25
     Bernie
    -0.25
     дизайн
    -0.24
    icap
    -0.24
    POSITIVE LOGITS
    åĿĹ
    0.28
    åĪĩ
    0.27
    cut
    0.27
    %M
    0.26
    yat
    0.25
     cut
    0.24
    vez
    0.24
     mol
    0.24
    è´¨
    0.24
    plot
    0.24
    Activations Density 0.010%

    No Known Activations