© 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. 19-TRANSCODER-HP
    7. 8012
    Prev
    Next
    INDEX
    Explanations

    leave

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Feinstein
    -0.27
    éĽħé»ij
    -0.27
    azen
    -0.26
     ваÑĢианÑĤ
    -0.25
     Preserve
    -0.24
     bÄĻdÄħ
    -0.24
    Ïİ
    -0.24
    æIJŀ好
    -0.23
    holding
    -0.23
    è¿Ļ款车
    -0.23
    POSITIVE LOGITS
    enced
    0.25
    ting
    0.25
     Maiden
    0.24
    èĭ¥æľī
    0.24
    _ng
    0.23
     Patt
    0.23
    ylon
    0.23
    æĭĽèĤ¡
    0.23
    éļĻ
    0.22
     altogether
    0.22
    Activations Density 0.339%

    No Known Activations