© 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. 63551
    Prev
    Next
    INDEX
    Explanations

    article snippets

    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
     quarters
    -0.27
     bastante
    -0.26
    èij©
    -0.26
    -serving
    -0.25
    ço
    -0.25
    umably
    -0.25
    quential
    -0.25
    send
    -0.24
    èĮļ
    -0.24
     svenska
    -0.24
    POSITIVE LOGITS
    rom
    0.30
    essim
    0.27
    orm
    0.27
    æ·±è¿ľ
    0.27
    hen
    0.26
    à¸ģาย
    0.26
    acht
    0.25
    ault
    0.25
    ana
    0.25
    _BUFF
    0.25
    Activations Density 16.238%

    No Known Activations