© 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. 32-TRANSCODER-HP
    7. 41607
    Prev
    Next
    INDEX
    Explanations

    vaccine

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    éli
    -0.31
    avern
    -0.26
    éģ¥è¿ľ
    -0.25
    adel
    -0.25
    /rem
    -0.24
     counselors
    -0.24
    æ°¸ä¹ħ
    -0.24
    elic
    -0.24
    èĸª
    -0.24
    aders
    -0.23
    POSITIVE LOGITS
    è¶ĭ
    0.30
    IFE
    0.29
    ĪæĿĥ
    0.28
    opsy
    0.27
    ÃŃo
    0.26
    ifs
    0.25
    phis
    0.25
    xiety
    0.25
    çļĦäºĭå®ŀ
    0.25
    vit
    0.24
    Activations Density 0.013%

    No Known Activations