© 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. 70862
    Prev
    Next
    INDEX
    Explanations

    rumors and speculation

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

    The neuron detects words signaling speculation or unverified reports (e.g. “rumours,” “gossip,” “reports,” “speculation”).

    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
    襟
    -0.94
    íc
    -0.94
    uchs
    -0.90
    fgaben
    -0.89
    дена
    -0.88
    分布
    -0.87
     εφ
    -0.87
     plaisir
    -0.86
     /*
    -0.85
     kullan
    -0.85
    POSITIVE LOGITS
     rumors
    1.69
     about
    1.54
     rumored
    1.49
     rumor
    1.45
     rumours
    1.34
     alleging
    1.28
     Rumors
    1.28
    ว่า
    1.28
     incluso
    1.21
     whispers
    1.20
    Activations Density 0.053%

    No Known Activations