© 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. 36605
    Prev
    Next
    INDEX
    Explanations

    Q followed by colon

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

    The neuron fires strongly on the very first significant token (word or symbol) at the start of each new text segment or paragraph.

    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
     Cualquier
    -0.97
     territ
    -0.96
    luft
    -0.96
    ﻷ
    -0.88
    ['_
    -0.88
     绘
    -0.88
    милия
    -0.87
     environnement
    -0.87
    éda
    -0.87
    варя
    -0.86
    POSITIVE LOGITS
     these
    1.06
     где
    0.88
     specific
    0.88
    Accueil
    0.84
     ecológico
    0.83
     suitable
    0.83
     where
    0.82
    ESTADO
    0.82
     climático
    0.82
     using
    0.80
    Activations Density 0.001%

    No Known Activations