© 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. 31-TRANSCODER-HP
    7. 19155
    Prev
    Next
    INDEX
    Explanations

    Research Institutions

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    deep
    -0.33
    æ·±
    -0.32
     deep
    -0.29
    æĸ½
    -0.29
    æ·±å±Ĥ
    -0.27
    éĤ¬
    -0.26
    æ·±åĪ»çļĦ
    -0.26
    æ·±ãģĦ
    -0.25
    ä¸ĢéĶ®
    -0.25
    ationToken
    -0.25
    POSITIVE LOGITS
     at
    0.35
     there
    0.34
    ed
    0.34
    Outside
    0.31
     (
    0.31
    s
    0.31
    's
    0.30
     outside
    0.29
     Outside
    0.29
    ,
    0.29
    Activations Density 0.888%

    No Known Activations