© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 30-TRANSCODER-HP
    7. 95337
    Prev
    Next
    INDEX
    Explanations

    genetically

    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
    Not in Any Lists

    No Comments

    Negative Logits
    Ñīи
    -0.28
     ``(
    -0.26
    allen
    -0.26
     ident
    -0.25
    æľīæľº
    -0.25
     cyclic
    -0.24
    ëĦĪ
    -0.24
     Orient
    -0.24
    hest
    -0.24
    .unpack
    -0.24
    POSITIVE LOGITS
     modification
    0.33
    ä¿®æĶ¹
    0.33
     modified
    0.30
    æĶ¹åıĺ
    0.29
    æĶ¹éĢł
    0.29
     modifier
    0.28
    åľ°æĸ¹
    0.27
     hybrid
    0.27
    ammed
    0.27
     erle
    0.27
    Activations Density 0.009%

    No Known Activations