© 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. 19-TRANSCODER-HP
    7. 8772
    Prev
    Next
    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Foot
    -0.29
    ç§ijæĬĢèĤ¡ä»½
    -0.26
    ç§ĺå¯Ĩ
    -0.25
    æī£
    -0.25
    èĽĭ
    -0.25
    çļĦç§ĺå¯Ĩ
    -0.23
    effect
    -0.23
    erus
    -0.22
    uddle
    -0.22
    доб
    -0.22
    POSITIVE LOGITS
    ddd
    0.27
    IMER
    0.26
    ported
    0.26
    CEL
    0.26
    psc
    0.26
    pute
    0.25
    ports
    0.24
    /mit
    0.24
    DDD
    0.24
    pla
    0.23
    Activations Density 0.017%

    No Known Activations