© 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. 34-GEMMASCOPE-RES-131K
    7. 92233
    Prev
    Next
    INDEX
    Explanations

    claims proven wrong

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

    The neuron strongly activates on first‐person narrative markers—especially the pronoun “I” (and occasional related short words like “was,” “boy,” “wrong”)—indicating it detects personal, confessional first‐person statements.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    ној
    -0.77
    род
    -0.73
     Recognition
    -0.72
    ἵ
    -0.71
     zdoby
    -0.69
    Zion
    -0.69
     HAN
    -0.68
    lix
    -0.68
     третий
    -0.68
    gans
    -0.68
    POSITIVE LOGITS
     wrong
    3.73
    wrong
    2.86
     Wrong
    2.75
    Wrong
    2.69
     WRONG
    2.61
    WRONG
    2.31
     proved
    2.30
     proven
    2.11
    错了
    2.05
     оши
    1.93
    Activations Density 0.035%

    No Known Activations