© 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. 112753
    Prev
    Next
    INDEX
    Explanations

    Units of time

    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
    tright
    -0.27
     Cory
    -0.25
    ç¿
    -0.25
    /view
    -0.25
    idor
    -0.24
    itals
    -0.24
    éĩįéĩį
    -0.23
    lifting
    -0.23
    åŁĻ
    -0.23
    edicine
    -0.23
    POSITIVE LOGITS
     Neal
    0.26
    (ele
    0.25
     bothered
    0.24
    å°¼äºļ
    0.24
    Blueprint
    0.24
    磨
    0.24
    NavigationView
    0.23
    Billy
    0.23
    remen
    0.23
    ä¸įä¸Ģæł·çļĦ
    0.23
    Activations Density 0.020%

    No Known Activations