© 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. 63182
    Prev
    Next
    INDEX
    Explanations

    2

    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
    -meta
    -0.27
    åīª
    -0.27
     ÐĽÐ¸
    -0.26
    åı¼
    -0.23
    åī²
    -0.23
    matcher
    -0.23
    å®ī举
    -0.23
    Anchor
    -0.23
     ann
    -0.23
     unloaded
    -0.23
    POSITIVE LOGITS
    stad
    0.29
    ;br
    0.27
    è®·
    0.25
    rypt
    0.24
    cum
    0.24
    è¿Ļ项
    0.24
    æ³ķ令
    0.24
    nar
    0.24
    layer
    0.23
    StrLn
    0.23
    Activations Density 1.489%

    No Known Activations