© 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. 27-TRANSCODER-HP
    7. 57277
    Prev
    Next
    INDEX
    Explanations

    a

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_27.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.27.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æĢ»åħ±
    -0.32
    满åĪĨ
    -0.31
    çļĦæ¯Ķä¾ĭ
    -0.29
    满
    -0.27
    åĽ½éģĵ
    -0.26
    Wil
    -0.26
     Wil
    -0.25
    .bel
    -0.25
    jur
    -0.25
    å¨ģ
    -0.25
    POSITIVE LOGITS
     idea
    0.30
    chosen
    0.28
    ç´§
    0.27
    ç·Ĭ
    0.27
     INNER
    0.27
    hetto
    0.27
     congest
    0.27
     notion
    0.27
    orie
    0.26
    onom
    0.26
    Activations Density 0.008%

    No Known Activations