© 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. 63733
    Prev
    Next
    INDEX
    Explanations

    Browser compatibility issues

    np_max-act · gemini-2.0-flash

    browser or website notifications that inform users about unsupported configurations (like disabled JavaScript, using an outdated browser, ad-blocker warnings, or guest access limitations).

    oai_token-act-pair · deepseek-v3Triggered by @finnwit

    Periods or commas indicating the end of sentences or pauses, often related to browser/technical issues or instructions.

    eleuther_acts_top20 · deepseek-v3Triggered by @finnwit
    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
    uese
    -0.32
     ama
    -0.31
    uet
    -0.29
    éķ¿
    -0.27
    éķĴ
    -0.26
    hei
    -0.26
    igner
    -0.26
    åĽłä¸ºä½ł
    -0.26
    æ¯ı次
    -0.25
    fü
    -0.25
    POSITIVE LOGITS
     {}.
    0.36
    azard
    0.28
     Wired
    0.27
     Rever
    0.26
    ramid
    0.26
     +%
    0.26
    ridge
    0.25
    /loader
    0.24
    Expired
    0.24
     Facing
    0.23
    Activations Density 5.810%

    No Known Activations