© 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. 31-TRANSCODER-HP
    7. 19594
    Prev
    Next
    INDEX
    Explanations

    uncommon words

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    stell
    -0.27
    纵åIJij
    -0.27
    Delta
    -0.26
    Multiplicity
    -0.26
    dling
    -0.26
    碱
    -0.26
    åѦç§ij
    -0.25
    å¹³çŃī
    -0.25
    è´¹
    -0.25
    æ³¼
    -0.25
    POSITIVE LOGITS
    arium
    0.25
     á»ķ
    0.25
    argin
    0.24
    outers
    0.24
    ä¸įæĢķ
    0.24
    åŀ¤
    0.24
     заб
    0.24
     endors
    0.23
    ampler
    0.23
    arlo
    0.23
    Activations Density 0.200%

    No Known Activations