© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 15859
    Prev
    Next
    INDEX
    Explanations

    past tense verbs

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

    The neuron strongly activates on tokens that consist of or contain numerical sequences (e.g. IDs, years, page numbers, codes, or other digit‐heavy strings).

    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
    batch
    0.88
    𝗂
    0.73
    upon
    0.71
     batch
    0.71
    зё
    0.71
     relinqu
    0.70
     Τα
    0.70
    拭
    0.70
     Это
    0.68
    ందా
    0.67
    POSITIVE LOGITS
     tempi
    0.80
    त्त
    0.78
     voies
    0.76
     tillsammans
    0.75
    tti
    0.74
     principais
    0.73
     offrent
    0.73
     offrant
    0.73
     differentiate
    0.70
     Haci
    0.70
    Activations Density 0.000%

    No Known Activations