© 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. 22-TRANSCODER-HP
    7. 112812
    Prev
    Next
    INDEX
    Explanations

    given

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    -Dec
    -0.26
    ç«ĭå¿Ĺ
    -0.25
    CAPE
    -0.25
    MEA
    -0.24
    /rem
    -0.24
    -API
    -0.24
    ä¸Ĭåīį
    -0.24
    -Nov
    -0.24
    pedia
    -0.24
    æĺİæľĹ
    -0.24
    POSITIVE LOGITS
    .listen
    0.29
     accom
    0.26
     accommodate
    0.26
     reflexivity
    0.25
     optionally
    0.24
     Yan
    0.24
     recreate
    0.24
    ÑĥÑĪ
    0.24
    ä»İèĢĮ使
    0.24
    uste
    0.23
    Activations Density 0.002%

    No Known Activations