© 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. 34-GEMMASCOPE-RES-131K
    7. 86933
    Prev
    Next
    INDEX
    Explanations

    is/was followed by descriptor

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

    The neuron is a “function‐word” detector—it lights up on common high‐frequency grammatical tokens (pronouns, auxiliaries, prepositions, conjunctions, etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    がありますが
    -0.84
     some
    -0.75
     all
    -0.75
    ⃨
    -0.74
     or
    -0.74
     through
    -0.73
    するの
    -0.73
     grito
    -0.72
    곽
    -0.72
     other
    -0.71
    POSITIVE LOGITS
    itself
    0.91
    бята
    0.89
    unzel
    0.88
    GMT
    0.87
     ถ้า
    0.86
    î
    0.85
     雷
    0.85
    …»
    0.85
    となっている
    0.84
    ologische
    0.84
    Activations Density 0.039%

    No Known Activations