© 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. 67334
    Prev
    Next
    INDEX
    Explanations

    words after punctuation

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

    The neuron is firing on tokens that begin a new segment or sentence—i.e. it marks sentence-initial words.

    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
     hice
    -0.98
     spod
    -0.95
    Sho
    -0.91
    İ
    -0.90
     pijama
    -0.88
    𝐼
    -0.88
    Ι
    -0.87
     añ
    -0.86
     kinderen
    -0.86
    </code>
    -0.85
    POSITIVE LOGITS
     enquired
    0.81
     المثال
    0.79
    倒是
    0.79
     které
    0.79
    idée
    0.77
     знать
    0.77
    Cannon
    0.76
    nextElement
    0.75
    aéroport
    0.75
     recently
    0.75
    Activations Density 0.007%

    No Known Activations