© 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. 150764
    Prev
    Next
    INDEX
    Explanations

    roughly

    np_max-act-logits · gemini-2.0-flash
    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
    -sur
    -0.28
    ogle
    -0.27
    æ´»åĬ¨
    -0.26
    åİ»åģļ
    -0.26
    NotNull
    -0.26
    _',
    -0.26
     Autor
    -0.26
    _goto
    -0.25
     stil
    -0.25
     Sty
    -0.25
    POSITIVE LOGITS
    iked
    0.27
    azzi
    0.26
    éħĬ
    0.26
    ugu
    0.25
    abo
    0.24
    ophile
    0.24
     Kidd
    0.24
    -hole
    0.24
    urent
    0.24
    ç»Ŀ
    0.24
    Activations Density 0.010%

    No Known Activations