© 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. 25-TRANSCODER-HP
    7. 108933
    Prev
    Next
    INDEX
    Explanations

    For

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    请大家
    -0.28
     Trustees
    -0.27
     syn
    -0.27
    uly
    -0.24
    build
    -0.24
    -build
    -0.24
     Devils
    -0.24
     sessions
    -0.24
     build
    -0.24
    .build
    -0.24
    POSITIVE LOGITS
     Watches
    0.31
    lopedia
    0.30
    .Skin
    0.30
    odo
    0.28
    furt
    0.28
    orado
    0.27
    AtPath
    0.27
    duk
    0.26
    éĻ¤
    0.26
    iband
    0.26
    Activations Density 1.033%

    No Known Activations