© 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. 84101
    Prev
    Next
    INDEX
    Explanations

    combining or combinations

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

    The neuron fires on numeric tokens, i.e. standalone numbers or numeric measurements (including decimals) in the text.

    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
    zené
    -0.85
     노
    -0.82
     موا
    -0.81
    ghal
    -0.80
    fromUtf
    -0.79
    ल्क
    -0.78
     baron
    -0.78
     dè
    -0.77
    шки
    -0.77
    ुप
    -0.75
    POSITIVE LOGITS
     combined
    2.72
     combination
    2.66
     combinations
    2.53
     combine
    2.38
     combining
    2.27
     combines
    2.20
    combined
    2.13
    combination
    2.09
     hybrid
    2.08
     COMBIN
    2.05
    Activations Density 0.147%

    No Known Activations