© 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. 32-TRANSCODER-HP
    7. 41949
    Prev
    Next
    INDEX
    Explanations

    Technical and research snippets

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    arest
    -0.28
    .Companion
    -0.28
    restart
    -0.26
    ained
    -0.26
    .finished
    -0.24
    PPER
    -0.24
    OwnProperty
    -0.24
    ç¼´è´¹
    -0.24
    .embedding
    -0.23
    çļĦä¸ĢéĿ¢
    -0.23
    POSITIVE LOGITS
    ile
    0.30
    asl
    0.30
    enth
    0.29
    itle
    0.28
    èģĬ
    0.26
    undi
    0.26
    osi
    0.25
    AS
    0.25
    秣
    0.25
    opy
    0.25
    Activations Density 0.000%

    No Known Activations