© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 6807
    Prev
    Next
    INDEX
    Explanations

    is

    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
    éĢļè¡Į
    -0.28
    ITT
    -0.26
    enton
    -0.25
     liá»ģn
    -0.25
    SenderId
    -0.25
     Minist
    -0.25
     entertaining
    -0.24
     Compatible
    -0.24
    aat
    -0.23
    ENTRY
    -0.23
    POSITIVE LOGITS
    ively
    0.29
    pest
    0.28
    isset
    0.27
    èĦ±è´«æĶ»åĿļ
    0.26
    cedure
    0.26
    pone
    0.26
    ķĮ
    0.25
    ivid
    0.25
    cedures
    0.24
    ched
    0.24
    Activations Density 0.006%

    No Known Activations