© 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. 41508
    Prev
    Next
    INDEX
    Explanations

    at

    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
    åijĬ
    -0.32
    åĻ¨
    -0.30
    æľº
    -0.27
    educt
    -0.27
     Gale
    -0.26
    ---
    -0.26
    ----
    -0.25
    ari
    -0.25
    ICENSE
    -0.25
     макÑģ
    -0.25
    POSITIVE LOGITS
     cro
    0.33
     watch
    0.26
     vä
    0.25
    æľīå¤ļ
    0.25
    åĩłæ¬¡
    0.25
    åĽºä½ĵ
    0.25
    ard
    0.25
    indr
    0.25
    é³ĸ
    0.24
    fäh
    0.24
    Activations Density 0.002%

    No Known Activations