© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 68733
    Prev
    Next
    INDEX
    Explanations

    expressing thoughts

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

    The neuron fires on hedging or opinion‐marker words—especially “think” (as in “I think”) and similar expressions of belief.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    ⎙
    -1.62
     Several
    -1.57
     "
    -1.56
     What
    -1.56
     and
    -1.55
     which
    -1.52
     Currently
    -1.48
     although
    -1.44
    icksburg
    -1.43
    tividade
    -1.42
    POSITIVE LOGITS
    ⵌ
    1.67
     your
    1.63
    個人的に
    1.56
     they
    1.53
    氿
    1.50
    ヤバ
    1.48
    ではでは
    1.47
     också
    1.46
    ふと
    1.46
    うわ
    1.45
    Activations Density 0.008%

    No Known Activations