© 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. 10962
    Prev
    Next
    INDEX
    Explanations

    cause numbers

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

    The neuron fires on standout content words (typically rarer, semantically important tokens) that often mark key statements or clauses.

    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
     มัน
    0.89
     ი
    0.88
    を感じ
    0.86
    ﯼ
    0.85
    ᱕
    0.80
     chúng
    0.80
     tutto
    0.79
     lane
    0.79
     satış
    0.79
    ında
    0.79
    POSITIVE LOGITS
    ANCE
    0.88
    ли
    0.75
     степен
    0.72
    م
    0.71
    ValArr
    0.70
    цем
    0.70
    ரிய
    0.69
    ностями
    0.65
    省略
    0.65
    ات
    0.65
    Activations Density 0.000%

    No Known Activations