© 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. 19436
    Prev
    Next
    INDEX
    Explanations

    code

    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
    inux
    -0.28
     por
    -0.27
    seau
    -0.27
    por
    -0.26
    çIJĨ
    -0.26
     POR
    -0.25
    debug
    -0.25
    roid
    -0.25
    :variables
    -0.25
    åĽ´
    -0.25
    POSITIVE LOGITS
    _HEAP
    0.29
    é¼ĵ
    0.28
    bolt
    0.28
    .ar
    0.27
    chet
    0.26
     barrels
    0.25
    edith
    0.25
    NotFoundError
    0.24
     TArray
    0.24
    tte
    0.24
    Activations Density 0.014%

    No Known Activations