© 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. 88964
    Prev
    Next
    INDEX
    Explanations

    for

    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
    Negative Logits
    mom
    -0.27
    athan
    -0.27
    çļĦåģļæ³ķ
    -0.26
    wor
    -0.26
    æľŁéĹ´
    -0.26
    çļĦæĸ¹æ³ķ
    -0.26
     Clar
    -0.26
    om
    -0.25
    çļĦæĥħåĨµ
    -0.25
     occupied
    -0.25
    POSITIVE LOGITS
    ','#
    0.27
    /licenses
    0.26
     <",
    0.26
    eling
    0.25
    _',
    0.24
    ?>↵
    0.24
    [];č↵
    0.23
    -distance
    0.23
    MEA
    0.23
    è¦ģåѦä¼ļ
    0.23
    Activations Density 0.003%

    No Known Activations