© 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. 22-TRANSCODER-HP
    7. 111895
    Prev
    Next
    INDEX
    Explanations

    encyclopedia

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ças
    -0.26
    lim
    -0.25
     papel
    -0.25
    ëıĮ
    -0.24
    slash
    -0.24
    cern
    -0.24
    çĥ¬
    -0.24
    æĬ¥ä»·
    -0.23
    (Cs
    -0.23
    åŀĭ
    -0.23
    POSITIVE LOGITS
    iens
    0.30
    иÑİ
    0.25
    /live
    0.24
    ä¸Ģè¡Į
    0.24
    -mounted
    0.24
     bada
    0.24
     ********************************************************
    0.23
    quirer
    0.23
    pter
    0.23
     powers
    0.23
    Activations Density 0.025%

    No Known Activations