© 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. 19370
    Prev
    Next
    INDEX
    Explanations

    biology/scientific contexts

    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
    .sy
    -0.27
    åĽ¾çĶ»
    -0.25
    .eval
    -0.25
    :red
    -0.25
     hạ
    -0.25
    _land
    -0.23
    abet
    -0.23
    паÑģ
    -0.23
     addCriterion
    -0.23
     Picture
    -0.23
    POSITIVE LOGITS
    cono
    0.28
    plate
    0.28
    ç¼®
    0.26
     ball
    0.26
    _mount
    0.26
     worm
    0.25
    ä½ĵ积
    0.25
    asaki
    0.24
    OCR
    0.24
     stock
    0.24
    Activations Density 0.085%

    No Known Activations