© 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. 20-TRANSCODER-HP
    7. 142298
    Prev
    Next
    INDEX
    Explanations

    religious texts using "for"

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    amat
    -0.27
     verw
    -0.25
    æľīä¸Ģå®ļ
    -0.25
    åħ´å¥ĭ
    -0.25
    vation
    -0.24
    è¿·æĥij
    -0.24
     herself
    -0.24
     privilege
    -0.24
    }_
    -0.24
    нÑİ
    -0.24
    POSITIVE LOGITS
    åħļç»Ħ
    0.27
    pg
    0.25
    ÙĤ
    0.25
    matic
    0.25
    ILES
    0.24
    -found
    0.24
     pii
    0.24
    临
    0.24
     Injectable
    0.24
     horn
    0.24
    Activations Density 11.534%

    No Known Activations