© 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. 18266
    Prev
    Next
    INDEX
    Explanations

    paid

    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
     past
    -0.37
    past
    -0.35
    Past
    -0.35
     Past
    -0.33
    è¿ĩåİ»
    -0.32
     Canyon
    -0.29
     Paste
    -0.28
    æĶ¯éĥ¨
    -0.25
    chen
    -0.25
    ева
    -0.24
    POSITIVE LOGITS
    åıijå±ķä¸Ń
    0.26
    illion
    0.26
    olu
    0.25
    Enums
    0.25
    []}
    0.24
    ungi
    0.24
    nullptr
    0.24
    ()],↵
    0.24
    æĢĢ
    0.24
    åīŁ
    0.23
    Activations Density 0.003%

    No Known Activations