© 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. 18186
    Prev
    Next
    INDEX
    Explanations

    about

    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
    çªģåĩºéĹ®é¢ĺ
    -0.32
    èĤ¤èī²
    -0.27
     nex
    -0.26
    PIO
    -0.26
    zure
    -0.25
    oulos
    -0.25
    redient
    -0.25
     preval
    -0.25
    permalink
    -0.24
     kino
    -0.24
    POSITIVE LOGITS
     ABOUT
    0.37
    about
    0.37
     about
    0.36
     About
    0.36
    About
    0.33
    _about
    0.33
    ABOUT
    0.31
     ANY
    0.29
    -about
    0.28
    篦
    0.27
    Activations Density 0.152%

    No Known Activations