© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130316
    Prev
    Next
    INDEX
    Explanations

    The neuron is picking up first-person commenter language (especially “I,” “my,” and “Thanks”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    positive sentiment and personal statements

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     السياس
    -0.97
    μως
    -0.97
    ּוֹ
    -0.95
     kepercayaan
    -0.91
     bringt
    -0.90
     अलग
    -0.88
    Trimethyl
    -0.88
    ঋ
    -0.88
    куда
    -0.88
    estructuras
    -0.87
    POSITIVE LOGITS
    Hoping
    1.12
     hoping
    1.03
    next
    1.02
     I
    1.02
     gostar
    1.01
    ıca
    0.96
     età
    0.94
     hope
    0.94
     hoped
    0.94
     descobri
    0.91
    Activations Density 0.011%

    No Known Activations