© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 111096
    Prev
    Next
    INDEX
    Explanations

    posted on

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron spikes on numeric and related metadata tokens (e.g. dates, post counts, quote/Originally labels, archive tags)—in other words, it detects forum‐style metadata elements.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    textual metadata from blogs/forums—dates, timestamps, “posted by” lines, quotes, and user/joined/post-count headers indicating post structure rather than content.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    line breaks or whitespace separating sections in blog posts and forum content.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     نمایید
    -0.75
     gonad
    -0.72
    Pues
    -0.69
    nasse
    -0.69
    nyt
    -0.69
    anceled
    -0.68
    贖
    -0.68
    bridal
    -0.68
    ={{
    -0.67
    Подробнее
    -0.66
    POSITIVE LOGITS
    Posted
    5.91
     Posted
    5.13
     posted
    4.78
    posted
    3.92
     posting
    3.58
     postings
    3.03
     Posting
    2.78
    posting
    2.67
    Posting
    2.64
     posts
    1.89
    Activations Density 0.203%

    No Known Activations