© 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. 32-TRANSCODER-HP
    7. 42466
    Prev
    Next
    INDEX
    Explanations

    Learn

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    åī©ä½Ļ
    -0.29
    áºŃm
    -0.27
    neo
    -0.27
    å®ī举
    -0.27
    gé
    -0.26
    ori
    -0.26
    ChildIndex
    -0.25
    rror
    -0.25
    åİ¢
    -0.25
    å¾Īå¤ļäºĭæĥħ
    -0.25
    POSITIVE LOGITS
     Online
    0.36
    ing
    0.35
     On
    0.33
     About
    0.31
     Det
    0.30
    edBy
    0.30
     Away
    0.29
    -ing
    0.29
    -On
    0.29
     To
    0.29
    Activations Density 0.002%

    No Known Activations