© 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. 24-TRANSCODER-HP
    7. 150932
    Prev
    Next
    INDEX
    Explanations

    Code or configuration files

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_24.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.24.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    spot
    -0.30
     clearly
    -0.27
     overhead
    -0.27
    istics
    -0.26
     unic
    -0.26
    常è¯Ĩ
    -0.26
    statt
    -0.26
    ling
    -0.25
    isten
    -0.25
     erre
    -0.25
    POSITIVE LOGITS
    -bo
    0.28
     sal
    0.26
    éĿ¢æĿ¿
    0.24
    ष
    0.24
    ej
    0.24
     Jones
    0.24
    å§Ķ书记
    0.24
    _CN
    0.24
     оÑĩеÑĢедÑĮ
    0.24
    æŁ¬
    0.24
    Activations Density 1.167%

    No Known Activations