© 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. 53887
    Prev
    Next
    INDEX
    Explanations

    of

    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
     norm
    -0.31
    èĩĨ
    -0.28
     ser
    -0.26
     Injector
    -0.26
     trace
    -0.26
    说èĩªå·±
    -0.26
    phet
    -0.25
     marg
    -0.25
     Hamilton
    -0.25
    Łèĥ½
    -0.25
    POSITIVE LOGITS
    Valor
    0.25
    åĩłå¼ł
    0.24
    è¿ĩ
    0.24
    Lights
    0.24
    å¹´å¹´åºķ
    0.24
     $?
    0.24
    åݻ年åIJĮæľŁ
    0.24
    .lastIndexOf
    0.23
    drivers
    0.23
    ël
    0.23
    Activations Density 0.019%

    No Known Activations