© 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. 22-TRANSCODER-HP
    7. 112388
    Prev
    Next
    INDEX
    Explanations

    Instructions and Explanations

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    -scal
    -0.25
    榻
    -0.25
    aktu
    -0.24
     Gilles
    -0.24
    ĵį
    -0.24
    éĹ·
    -0.24
    ffect
    -0.24
    åĬ¨ä½ľ
    -0.24
    æĵįä½ľ
    -0.23
    èIJ½
    -0.23
    POSITIVE LOGITS
     Chambers
    0.30
    asters
    0.30
    _NC
    0.27
    zig
    0.25
    âĶĢ
    0.24
     Tomb
    0.24
    ?(:
    0.24
    åIJĮæł·æĺ¯
    0.24
     Silent
    0.24
    áµĴ
    0.23
    Activations Density 0.002%

    No Known Activations