© 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. 75090
    Prev
    Next
    INDEX
    Explanations

    still, continue, maintain

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

    The neuron fires on first‐person self‐references—especially “I” plus verbs expressing personal abilities, desires, or ongoing actions.

    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
     qualsiasi
    -1.05
     cualquier
    -1.03
     each
    -1.00
     älteren
    -0.94
     qualunque
    -0.91
     cualquiera
    -0.90
     ciasc
    -0.89
     любого
    -0.88
     persönlichen
    -0.88
     betrieb
    -0.87
    POSITIVE LOGITS
     still
    1.78
    still
    1.41
     tetap
    1.36
     yine
    1.27
     vẫn
    1.18
    Still
    1.16
    还是
    1.11
    continue
    1.10
     Nederlandse
    1.08
     nadal
    1.05
    Activations Density 0.060%

    No Known Activations