© 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. 6635
    Prev
    Next
    INDEX
    Explanations

    stdio

    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
    ÑģÑĤвÑĥ
    -0.27
    neh
    -0.25
    apses
    -0.25
    lds
    -0.24
    _queries
    -0.24
    aneously
    -0.23
    ableView
    -0.23
    umba
    -0.23
    quent
    -0.23
    antas
    -0.23
    POSITIVE LOGITS
    ä¸Ĭåij¨
    0.28
    没ä»Ģä¹Ī
    0.28
    Weekly
    0.27
    廪
    0.27
     Weekly
    0.27
    otto
    0.27
    è½®å»ĵ
    0.25
    lex
    0.25
    èĢĹ
    0.25
     semana
    0.25
    Activations Density 0.054%

    No Known Activations