© 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. 30-TRANSCODER-HP
    7. 95133
    Prev
    Next
    INDEX
    Explanations

    years and numbers

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     least
    -0.28
    grave
    -0.26
     Least
    -0.26
    çļĦæīĭ
    -0.26
    alto
    -0.25
    æĬĬæīĭ
    -0.25
    æĮij
    -0.24
    è·³è·ĥ
    -0.24
     extrem
    -0.24
    _planes
    -0.24
    POSITIVE LOGITS
    obs
    0.29
    ?type
    0.27
    PWD
    0.25
    Unsigned
    0.25
    fon
    0.25
    'L
    0.25
    ÑĦон
    0.24
    æĴ¬
    0.24
    æľĽåİ»
    0.24
     Hale
    0.24
    Activations Density 0.897%

    No Known Activations