© 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. 31-TRANSCODER-HP
    7. 18427
    Prev
    Next
    INDEX
    Explanations

    Geography and landscapes

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    itors
    -0.29
    é¦Ĩ
    -0.27
     Homeland
    -0.27
    åѸéĻ¢
    -0.26
    utors
    -0.25
     PSG
    -0.25
    ä¸Ģç§Ĵ
    -0.25
     honors
    -0.25
    åѦéĻ¢
    -0.25
     institute
    -0.25
    POSITIVE LOGITS
    å¼ĢæĶ¾
    0.33
     little
    0.33
    å¹³åĿ¦
    0.33
     terr
    0.32
     sand
    0.32
     open
    0.31
     rocky
    0.30
    ering
    0.30
    eras
    0.30
     dense
    0.30
    Activations Density 2.003%

    No Known Activations