© 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. 33-TRANSCODER-HP
    7. 54025
    Prev
    Next
    INDEX
    Explanations

    air

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    è¯ķçĿĢ
    -0.30
     Pipes
    -0.29
     pipe
    -0.28
    omm
    -0.27
    pipe
    -0.27
    omon
    -0.26
    çŃ·åŃIJ
    -0.26
    æĿĵ
    -0.26
    omite
    -0.26
    utions
    -0.26
    POSITIVE LOGITS
     acum
    0.27
     steam
    0.26
    tes
    0.26
    æIJ¬
    0.26
    ::_
    0.26
    å®ıè§Ĥç»ıæµİ
    0.25
     NAV
    0.25
    aza
    0.25
    ãģĦãģĨ
    0.25
    //-----------------------------------------------------------------------------↵
    0.25
    Activations Density 0.008%

    No Known Activations