© 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. 30-TRANSCODER-HP
    7. 95595
    Prev
    Next
    INDEX
    Explanations

    ish

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    AFF
    -0.29
    .exports
    -0.28
    hall
    -0.26
     Instit
    -0.26
    _commit
    -0.25
    aff
    -0.25
    _COMMIT
    -0.24
    çłĶ
    -0.24
    _CONTROLLER
    -0.24
     advocates
    -0.23
    POSITIVE LOGITS
    inan
    0.30
     Gab
    0.29
     Coleman
    0.27
    -assets
    0.26
     CActive
    0.25
    ç§ijæķĻ
    0.25
     ctype
    0.24
    ä¸įåı¯éģ¿åħį
    0.24
    ogne
    0.24
    群
    0.24
    Activations Density 0.100%

    No Known Activations