© 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. 7157
    Prev
    Next
    INDEX
    Explanations

    variations and types

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

    This neuron fires most strongly on BPE tokens that constitute complete words (i.e. standalone word pieces rather than fragments).

    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
     amplo
    0.84
     Excellent
    0.79
    },$
    0.77
     realizado
    0.73
    urada
    0.73
     defesa
    0.73
     Excelente
    0.73
     excelente
    0.71
     नर्स
    0.71
     기대
    0.71
    POSITIVE LOGITS
    ר
    0.79
    hate
    0.78
     suitors
    0.78
    ца
    0.76
    عه
    0.72
    vice
    0.71
    carte
    0.71
    పాటు
    0.71
    لې
    0.71
    र
    0.71
    Activations Density 0.000%

    No Known Activations