© 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. 15890
    Prev
    Next
    INDEX
    Explanations

    input validation messages

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

    The neuron consistently lights up on the very last content word before each segment break, i.e. it detects the final token of a text block.

    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
    сний
    0.73
    ق
    0.73
    statistic
    0.68
    notin
    0.68
    Choosing
    0.68
    هو
    0.66
    тування
    0.66
    是真的
    0.65
    ခဲ့
    0.64
    traits
    0.64
    POSITIVE LOGITS
     ener
    0.88
     razão
    0.88
    ଞ
    0.83
     ды
    0.82
     Эд
    0.81
     depósito
    0.81
    ремя
    0.80
     παν
    0.80
    endim
    0.80
    ര്‍ന്ന്
    0.80
    Activations Density 0.000%

    No Known Activations