© 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. 18880
    Prev
    Next
    INDEX
    Explanations

    Names and locations

    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
     gu
    -0.30
     Coun
    -0.29
    guard
    -0.27
    件
    -0.27
    æ§ģ
    -0.27
    gü
    -0.26
    /sw
    -0.25
    .guard
    -0.25
    aut
    -0.24
    alto
    -0.24
    POSITIVE LOGITS
    åĿļå®ļä¸į移
    0.30
    ystery
    0.25
    å¼ĭ
    0.25
    ole
    0.25
     perror
    0.25
    posites
    0.25
    .scala
    0.25
    \brief
    0.24
     Nan
    0.24
    abis
    0.24
    Activations Density 0.421%

    No Known Activations