© 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. 31-TRANSCODER-HP
    7. 18776
    Prev
    Next
    INDEX
    Explanations

    programming control functions

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    yps
    -0.30
    çĴľ
    -0.28
    myp
    -0.28
    é»į
    -0.26
    èij¡
    -0.26
    çĻ»
    -0.26
    zen
    -0.25
     kup
    -0.25
    æ¦ľ
    -0.25
    åľĴ
    -0.24
    POSITIVE LOGITS
     ctrl
    0.74
     contr
    0.70
    contr
    0.69
     control
    0.67
    æݧåĪ¶
    0.67
     controller
    0.66
     Ctrl
    0.66
    ctrl
    0.66
     controllers
    0.65
     contro
    0.65
    Activations Density 22.564%

    No Known Activations