© 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. 20-TRANSCODER-HP
    7. 108186
    Prev
    Next
    INDEX
    Explanations

    Discourse markers

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    _("
    -0.27
    atri
    -0.26
    aling
    -0.26
    epy
    -0.26
    iferay
    -0.24
     Peters
    -0.24
    interop
    -0.24
    å·§åIJĪ
    -0.24
     privilege
    -0.24
     pers
    -0.23
    POSITIVE LOGITS
    ÙĬÙĦØ©
    0.30
    èļ¤
    0.27
    apes
    0.27
    æĹ©æĻļ
    0.27
     there
    0.26
    _chip
    0.24
    åħĪè¿Ľåįķä½į
    0.24
    sap
    0.24
    ä¸į论æĺ¯
    0.24
     VH
    0.24
    Activations Density 26.616%

    No Known Activations