© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 31-TRANSCODER-HP
    7. 19950
    Prev
    Next
    INDEX
    Explanations

    miscellaneous text snippets

    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
    Negative Logits
    obox
    -0.27
    è¾ħ
    -0.25
    ission
    -0.25
    åıĤè°ĭ
    -0.25
    iais
    -0.25
    bean
    -0.24
    FILE
    -0.24
     picker
    -0.24
    beta
    -0.24
    hoot
    -0.23
    POSITIVE LOGITS
     tow
    0.33
    utom
    0.29
    amac
    0.28
    иÑĤ
    0.28
    иÑĤа
    0.27
    itant
    0.26
    amus
    0.26
    ir
    0.26
    otas
    0.25
    itat
    0.25
    Activations Density 34.038%

    No Known Activations