© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 130997
    Prev
    Next
    INDEX
    Explanations

    considering, given

    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
    Negative Logits
    Safety
    -0.25
    åľ¨ç½ij绾
    -0.25
    æĥ³è±¡åĬĽ
    -0.24
     Eaton
    -0.24
     bem
    -0.24
    wit
    -0.24
     cuz
    -0.23
    æľīåħ³
    -0.23
    .getServer
    -0.23
    ãģķ
    -0.23
    POSITIVE LOGITS
    bian
    0.27
    /lic
    0.26
    oen
    0.25
     Cave
    0.25
     Div
    0.25
    èĢģ家
    0.25
    pend
    0.24
     buff
    0.24
     pend
    0.24
    à¸ģรรม
    0.24
    Activations Density 19.057%

    No Known Activations