© 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. 142164
    Prev
    Next
    INDEX
    Explanations

    separators

    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
     jednoczeÅĽnie
    -0.27
    -yyyy
    -0.26
    ç»ĵæĿŁ
    -0.26
     pun
    -0.25
    _decorator
    -0.24
    å²Ĥ
    -0.24
    é²ľèĬ±
    -0.24
    życz
    -0.24
     printers
    -0.24
    _pause
    -0.23
    POSITIVE LOGITS
    itics
    0.29
    bern
    0.28
    inea
    0.27
     tour
    0.27
    usted
    0.26
     guard
    0.25
    rames
    0.23
    -operator
    0.23
    cth
    0.23
    eso
    0.23
    Activations Density 9.966%

    No Known Activations