© 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. 22-GEMMASCOPE-RES-131K
    7. 527
    Prev
    Next
    INDEX
    Explanations

    modal verbs of possibility

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

    The neuron flags speculative or modal expressions—words that signal possibility or uncertainty (e.g. “may,” “might,” “possibly”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    は約
    -1.16
    個室
    -1.07
    amma
    -0.98
    ポイントは
    -0.98
     probably
    -0.91
    sprüng
    -0.89
    bàn
    -0.88
    pielt
    -0.88
     iconLine
    -0.88
     مشابه
    -0.87
    POSITIVE LOGITS
    mpy
    1.19
     different
    1.16
     różne
    1.06
     több
    1.05
     différentes
    1.05
    𝗥
    1.03
    ₦
    1.02
     multiple
    1.00
    zeria
    0.99
    多种
    0.99
    Activations Density 0.060%

    No Known Activations