© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 64048
    Prev
    Next
    INDEX
    Explanations

    numbers followed by thousand or k

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

    The neuron is detecting numeric tokens or digit sequences (numbers).

    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
     compression
    -0.79
    Compression
    -0.77
     Compression
    -0.75
     Lager
    -0.73
     пові
    -0.67
    cett
    -0.67
    compression
    -0.66
    である
    -0.66
    ésultat
    -0.65
     Vast
    -0.63
    POSITIVE LOGITS
     thousand
    1.28
    thousand
    0.96
     Thousand
    0.95
     тыс
    0.89
     hundred
    0.88
     tisíc
    0.88
    Thousand
    0.84
    oit
    0.81
     ribu
    0.81
     هزار
    0.81
    Activations Density 0.031%

    No Known Activations