© 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. 141944
    Prev
    Next
    INDEX
    Explanations

    invasive

    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
    主人
    -0.28
    äºĶè¡Į
    -0.25
    INESS
    -0.23
    verture
    -0.23
    _queries
    -0.23
    PLUGIN
    -0.23
    èĮĮ
    -0.23
    agenta
    -0.23
    PCM
    -0.23
     champions
    -0.23
    POSITIVE LOGITS
    /exp
    0.29
    .Exp
    0.26
    ervoir
    0.26
     tens
    0.25
    mente
    0.25
     whether
    0.25
    cou
    0.25
    bek
    0.24
    bsub
    0.23
    ös
    0.23
    Activations Density 6.649%

    No Known Activations