© 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. 12-TRANSCODER-HP
    7. 21967
    Prev
    Next
    INDEX
    Explanations

    share

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_12.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.12.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    ibo
    -0.29
    æ·Ģ
    -0.28
    rint
    -0.26
    æ·±
    -0.25
    igg
    -0.25
    igar
    -0.25
    linger
    -0.24
    urry
    -0.24
    URY
    -0.24
     ÄIJình
    -0.24
    POSITIVE LOGITS
    :convert
    0.29
    --)↵
    0.26
     spelled
    0.26
    .gameserver
    0.26
    teness
    0.26
    èĥ½åĬĽåĴĮ
    0.25
     IPA
    0.25
    íĨ±
    0.25
     Iss
    0.25
    åIJĦæĸ¹
    0.25
    Activations Density 0.005%

    No Known Activations