© 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. 12-TRANSCODER-HP
    7. 22876
    Prev
    Next
    INDEX
    Explanations

    Code, programming

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_12.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.12.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ä¼ł
    -0.27
    .ACCESS
    -0.26
    grounds
    -0.26
    AMES
    -0.26
    ground
    -0.26
    æ°´åĪĨ
    -0.26
     intervened
    -0.25
     sense
    -0.25
     liqu
    -0.25
     polls
    -0.25
    POSITIVE LOGITS
    é¢Ĩ导èĢħ
    0.29
    bane
    0.27
    =>"
    0.25
    ä¸īæľŁ
    0.25
    á¿Ĩ
    0.24
    ître
    0.24
    éļ¾
    0.24
    éĶģå®ļ
    0.24
    åĩłç§į
    0.24
    'gc
    0.23
    Activations Density 0.001%

    No Known Activations