© 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. 107113
    Prev
    Next
    INDEX
    Explanations

    Organizations

    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
    eways
    -0.26
    _construct
    -0.25
    alez
    -0.25
    azes
    -0.25
    èĩ§
    -0.25
     maj
    -0.24
    odo
    -0.24
    etas
    -0.24
    åĬ¡
    -0.24
    aze
    -0.23
    POSITIVE LOGITS
    ÃĶ
    0.25
    Bias
    0.24
    çħ§æĺİ
    0.24
    æĽĿåħī
    0.23
    åĪĨåĮº
    0.23
     Fix
    0.23
     breakup
    0.23
    ÑģÑĭ
    0.23
     Irving
    0.23
     Indices
    0.23
    Activations Density 8.495%

    No Known Activations