© 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. 45742
    Prev
    Next
    INDEX
    Explanations

    sentence endings and subsequent words

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

    The neuron fires for numeric tokens (multi‐digit numbers, e.g. years, measurements, quantities).

    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
     again
    -1.22
     Again
    -1.16
    },
    -1.16
    Again
    -1.11
    ​,
    -1.01
     питання
    -0.99
     Schlafzimmer
    -0.97
    again
    -0.95
    지만
    -0.93
     ‐
    -0.92
    POSITIVE LOGITS
    Показать
    1.87
    ↵↵↵↵↵↵↵
    1.80
    ↵↵↵↵↵↵↵↵↵
    1.74
    ↵↵↵↵
    1.73
    ↵↵↵↵↵
    1.72
    ↵↵↵↵↵↵
    1.66
    ↵↵↵
    1.63
    ↵↵↵↵↵↵↵↵
    1.59
    ↵↵↵↵↵↵↵↵↵↵
    1.58
    ↵↵↵↵↵↵↵↵↵↵↵↵↵
    1.43
    Activations Density 0.154%

    No Known Activations