© 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. 34-GEMMASCOPE-RES-131K
    4. 11910
    Prev
    Next
    INDEX
    Explanations

    twilight and night sky

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

    The neuron fires on sentence-boundary cues—i.e. words and punctuation that mark the starts or ends of sentences.

    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
    rotter
    -0.79
    argas
    -0.78
    äsident
    -0.76
    虐
    -0.76
    kinase
    -0.75
    ungeon
    -0.75
    tifikat
    -0.75
    lation
    -0.75
    メント
    -0.74
    finement
    -0.74
    POSITIVE LOGITS
     instrumental
    0.91
     twilight
    0.83
     sky
    0.83
     nights
    0.83
     Instrumental
    0.81
     tell
    0.79
     multiplicative
    0.79
     background
    0.78
     nightly
    0.78
     calibration
    0.76
    Activations Density 0.012%

    No Known Activations