© 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. 53945
    Prev
    Next
    INDEX
    Explanations

    capture/catch

    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
    WARE
    -0.28
    æĮĤçīĮ
    -0.27
    çIJĽ
    -0.25
     NR
    -0.24
    é½IJ
    -0.24
     Copp
    -0.24
    жд
    -0.24
     Cf
    -0.24
    .mac
    -0.24
    awan
    -0.24
    POSITIVE LOGITS
     capture
    1.08
     captured
    1.07
     capturing
    1.05
    æįķ
    1.02
     captures
    1.02
     catch
    1.01
     catching
    0.94
     caught
    0.94
     Capture
    0.93
     catches
    0.92
    Activations Density 0.291%

    No Known Activations