© 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. 142112
    Prev
    Next
    INDEX
    Explanations

    Future tense/predictions

    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
    ework
    -0.35
    Foot
    -0.26
    /static
    -0.25
    amo
    -0.25
    aux
    -0.24
     Purs
    -0.24
    _deinit
    -0.24
     Annotation
    -0.23
    .vm
    -0.23
    quiz
    -0.23
    POSITIVE LOGITS
    åĶ¤
    0.28
    ç«ĸ
    0.27
    æĿ¥çļĦ
    0.25
    ivable
    0.25
    ês
    0.24
    RYPT
    0.24
    pon
    0.24
    æĢİä¹Īèĥ½
    0.23
    ifact
    0.23
    :;↵
    0.23
    Activations Density 17.270%

    No Known Activations