© 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. 25-TRANSCODER-HP
    7. 108480
    Prev
    Next
    INDEX
    Explanations

    care

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    çİīçŁ³
    -0.27
    Re
    -0.25
     Studies
    -0.25
     weights
    -0.25
    ÑĢаб
    -0.24
     greed
    -0.24
    ä¾Ľ
    -0.24
    éĢļè¡Į
    -0.24
    é¾Ļ头
    -0.24
    .FindControl
    -0.24
    POSITIVE LOGITS
    yük
    0.30
     Huck
    0.29
    enci
    0.25
    ASON
    0.25
    åŁİéĩĮ
    0.24
    åľºåĿĩ
    0.23
    bbe
    0.23
    овал
    0.23
    eren
    0.23
    poons
    0.23
    Activations Density 0.972%

    No Known Activations