© 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. 12108
    Prev
    Next
    INDEX
    Explanations

    father's father

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

    This neuron selectively fires on relatively rare, content-bearing tokens—typically longer or more information-dense words that have high TF–IDF weight in the document.

    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
     quedan
    0.85
     muit
    0.80
     quede
    0.79
     nadie
    0.77
     emisión
    0.77
    рона
    0.76
     мел
    0.75
    𝚘
    0.74
     lineal
    0.73
     quedando
    0.73
    POSITIVE LOGITS
    ן
    0.89
    ′
    0.84
    ’
    0.79
    ین
    0.79
    indungi
    0.77
    اني
    0.75
     rejoice
    0.75
    ategory
    0.74
     probiotics
    0.74
    putnik
    0.73
    Activations Density 0.000%

    No Known Activations