© 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-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 16312
    Prev
    Next
    INDEX
    Explanations

    combines or combines sentences

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

    The neuron fires on numeric tokens—especially multi‐digit numbers and decimal figures—i.e. it detects quantitative measurements or statistics.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     l
    0.82
    ्स
    0.77
     Tren
    0.74
     Kl
    0.74
     kristall
    0.73
     klar
    0.70
     Alright
    0.70
     Bent
    0.70
     ura
    0.70
     afgelopen
    0.70
    POSITIVE LOGITS
    Spider
    0.87
    чні
    0.84
    Statistic
    0.82
    म
    0.80
    Statistics
    0.79
    Houses
    0.79
    Security
    0.79
    Convolution
    0.77
    стное
    0.76
    стные
    0.75
    Activations Density 0.000%

    No Known Activations