© 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. 150709
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    ä¸ĩåįĥ
    -0.30
    [word
    -0.28
    樵
    -0.28
    ello
    -0.28
    makers
    -0.27
    篮
    -0.26
    esson
    -0.25
     [[[
    -0.25
    æ¡Ĩ
    -0.25
    ball
    -0.24
    POSITIVE LOGITS
     undercut
    0.25
    yr
    0.24
     exacerb
    0.24
     recreational
    0.24
    åıijå±ķä¸Ń
    0.24
    ÃŃcio
    0.23
    弥补
    0.23
     preferably
    0.22
    ác
    0.22
     deÄŁil
    0.22
    Activations Density 0.131%

    No Known Activations

    This feature has no known activations.