© 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. 87145
    Prev
    Next
    INDEX
    Explanations

    prepositions

    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
     titleLabel
    -0.26
    iedo
    -0.26
     bubbles
    -0.25
    ãģªãģĦãĤĪãģĨãģ«
    -0.25
     Tee
    -0.25
     Allied
    -0.24
    ampions
    -0.24
     Comm
    -0.24
     ná»ģn
    -0.23
    åĮ£
    -0.23
    POSITIVE LOGITS
     each
    0.82
    each
    0.81
    æ¯ı个
    0.73
    æ¯ıä¸Ģ
    0.68
    Each
    0.68
     Each
    0.65
     EACH
    0.65
    æ¯ı
    0.63
     chaque
    0.60
    æ¯ıä¸Ģ个
    0.60
    Activations Density 0.016%

    No Known Activations