© 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. 10-GEMMASCOPE-RES-131K
    7. 42814
    Prev
    Next
    INDEX
    Explanations

    at the risk of

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

    The neuron fires on hedging or disclaimer phrases—e.g. “no offence,” “at the risk of,” “don’t want to,” “hope,” “jinx myself,” etc.—marking polite mitigations or caveats.

    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
     Its
    -1.16
    したこと
    -1.14
    と思ったら
    -1.12
    会話
    -1.09
     decided
    -1.09
    ll
    -1.09
    日至
    -1.09
    <h3>
    -1.08
     for
    -1.07
     deprived
    -1.07
    POSITIVE LOGITS
     '
    1.56
     تمامی
    1.45
     SqlCommand
    1.43
     становятся
    1.41
    悖
    1.40
     salió
    1.38
     '_
    1.38
     '#
    1.38
     balcón
    1.38
    chartInstance
    1.34
    Activations Density 0.018%

    No Known Activations