© 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. 96955
    Prev
    Next
    INDEX
    Explanations

    resistant

    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
    éĺµ
    -0.28
     scramble
    -0.27
    æĸ°å½¢åĬ¿ä¸ĭ
    -0.26
    |array
    -0.25
     evolution
    -0.25
    ç¨Ģ
    -0.25
     pardon
    -0.25
    å¸ı
    -0.25
     flats
    -0.25
    collections
    -0.25
    POSITIVE LOGITS
    listed
    0.29
    variant
    0.28
    lasting
    0.27
    çīĪæľ¬
    0.26
     variants
    0.26
    abler
    0.26
    /group
    0.26
    abh
    0.26
    _variant
    0.25
    infra
    0.25
    Activations Density 0.888%

    No Known Activations