© 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. 42133
    Prev
    Next
    INDEX
    Explanations

    sub

    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
     Cont
    -0.24
     sh
    -0.24
     Guess
    -0.24
    -cn
    -0.24
    æ¯Ķä¾ĭ
    -0.24
    agan
    -0.24
    aper
    -0.23
    主管éĥ¨éŨ
    -0.23
     Manit
    -0.23
     Combine
    -0.23
    POSITIVE LOGITS
    verted
    0.40
    ersen
    0.30
    vert
    0.30
    erland
    0.28
    åħ¶æ¬¡æĺ¯
    0.28
    plementation
    0.28
    MLElement
    0.27
    inker
    0.26
    tractive
    0.26
    erde
    0.26
    Activations Density 0.013%

    No Known Activations