© 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. 31-TRANSCODER-HP
    7. 18588
    Prev
    Next
    INDEX
    Explanations

    Code and inventions

    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
    å¢ĥ
    -0.36
    大éĻĨ
    -0.29
     Boys
    -0.25
    eah
    -0.25
    stab
    -0.25
    etta
    -0.25
    OUN
    -0.25
    mans
    -0.24
     Competitive
    -0.24
     boys
    -0.24
    POSITIVE LOGITS
     bó
    0.26
    è¿¢
    0.26
    æ¼Ĥ
    0.25
    ÃŃv
    0.25
    uga
    0.25
     imp
    0.24
    æķ´é¡¿
    0.24
    urar
    0.24
    ertools
    0.24
    éģİåİ»
    0.24
    Activations Density 0.096%

    No Known Activations