© 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. 33-TRANSCODER-HP
    7. 54271
    Prev
    Next
    INDEX
    Explanations

    say number 3

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ä»ĸ
    -0.28
     Shot
    -0.26
     h
    -0.26
     hÃłnh
    -0.25
     Subscribe
    -0.25
     tes
    -0.25
     g
    -0.24
     He
    -0.23
    é²ľ
    -0.23
     history
    -0.23
    POSITIVE LOGITS
    opolitan
    0.26
    indi
    0.25
    èij¡
    0.25
    æł·æĿ¿
    0.25
    ogi
    0.25
     Phase
    0.24
    __,__
    0.24
    à¸ĵ
    0.24
    çıŃåŃIJ
    0.23
     unborn
    0.23
    Activations Density 0.030%

    No Known Activations