© 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. 16335
    Prev
    Next
    INDEX
    Explanations

    forms of harm

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

    The neuron fires on archaic (Early Modern English) present‐tense verb forms ending in “-eth.”

    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
    Lastly
    0.76
     voluminous
    0.72
    ∞</
    0.71
    
    0.70
     contentious
    0.68
    ())->
    0.68
    当該
    0.68
     nombrado
    0.67
    আছে
    0.66
     অভ্য
    0.66
    POSITIVE LOGITS
     Jans
    0.84
    m
    0.79
     nahin
    0.78
     Jel
    0.76
    arie
    0.74
    y
    0.74
    erts
    0.73
    yawa
    0.73
    ません
    0.72
     Bukan
    0.72
    Activations Density 0.001%

    No Known Activations