© 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. 68112
    Prev
    Next
    INDEX
    Explanations

    start of new phrases

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

    The neuron is detecting sentence-ending punctuation (marks like “.”, “?” and “!”).

    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
    ांस
    -1.03
     senaste
    -0.98
     flesta
    -0.87
     ziemi
    -0.85
     öpp
    -0.83
     implantation
    -0.81
    焦虑
    -0.81
     Jahrhunder
    -0.80
     thier
    -0.77
    PlayerPrefs
    -0.77
    POSITIVE LOGITS
     pusi
    1.06
    Again
    1.04
     again
    0.98
     custodia
    0.96
     Again
    0.96
     قاشق
    0.95
    yship
    0.93
    These
    0.91
     fisio
    0.90
    民宿
    0.90
    Activations Density 0.001%

    No Known Activations