© 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. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 86506
    Prev
    Next
    INDEX
    Explanations

    presenting information

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

    The neuron detects self-referential framing or introduction phrases that announce the article’s purpose or audience (e.g. “We will be sharing…,” “This article is for…,” “This post is an endeavour to…”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     at
    -1.84
     through
    -1.66
    жды
    -1.64
    ného
    -1.57
    星空
    -1.52
     four
    -1.52
     during
    -1.52
     in
    -1.46
    蔑
    -1.45
     culprits
    -1.42
    POSITIVE LOGITS
     you
    2.19
     constater
    1.53
     your
    1.45
     potrebbe
    1.42
    艄
    1.41
     corriger
    1.39
     groote
    1.38
    ᐈ
    1.38
     who
    1.37
    あなたが
    1.37
    Activations Density 0.116%

    No Known Activations