© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 35807
    Prev
    Next
    INDEX
    Explanations

    pronouns and question words

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

    The neuron is strongly triggered by line‐break boundaries and the very first tokens (including punctuation or common pronouns) at the start of each new line or block.

    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
    itosti
    -1.16
     выполнить
    -1.13
     although
    -1.07
     ответить
    -1.01
    ~ 
    -1.00
     приобрести
    -0.98
     mainly
    -0.98
    انواع
    -0.98
     possibly
    -0.98
     probably
    -0.98
    POSITIVE LOGITS
     vijf
    1.10
    ovým
    1.09
     Высо
    1.08
     seleccionados
    1.07
     costos
    1.02
     cyl
    0.99
     มี
    0.98
    的眼神
    0.97
     ребенка
    0.97
    ällä
    0.97
    Activations Density 0.023%

    No Known Activations