© 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. 12-TRANSCODER-HP
    7. 23196
    Prev
    Next
    INDEX
    Explanations

    computer code and technical jargon

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_12.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.12.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Sinai
    -0.28
     пÑĢÑı
    -0.28
     totaled
    -0.26
     Clair
    -0.25
    cntl
    -0.25
    çŃ·åŃIJ
    -0.25
     endPoint
    -0.24
    esting
    -0.24
    ersonic
    -0.24
    æѦæŀĹ
    -0.24
    POSITIVE LOGITS
    AndPassword
    0.27
    Ụ
    0.26
    å®ŀçİ°
    0.26
    çļĦåħ·ä½ĵ
    0.26
    æ·±åĪ»
    0.26
    èĥĮæĻ¯
    0.26
    éĺħ读
    0.25
    çļĦå®ŀéĻħ
    0.25
    VML
    0.25
    æݨè¡Į
    0.25
    Activations Density 0.016%

    No Known Activations