© 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. 107191
    Prev
    Next
    INDEX
    Explanations

    arguments and opinions

    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
    æľŁæľĽ
    -0.27
    Normalization
    -0.27
    scar
    -0.26
    å¸ĮæľĽèĩªå·±
    -0.26
    _friend
    -0.26
    prop
    -0.25
    Retention
    -0.25
    Clock
    -0.25
    ç»ĻåĪ«äºº
    -0.24
     clocks
    -0.24
    POSITIVE LOGITS
    ÅŁÄ±
    0.28
    ersive
    0.27
    č↵        č↵
    0.26
    éĻµ
    0.26
    cors
    0.26
    éģ¿
    0.25
    etus
    0.25
    èĥĮéĿ¢
    0.25
    ç²¥
    0.25
    樯
    0.25
    Activations Density 16.797%

    No Known Activations