© 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. 12765
    Prev
    Next
    INDEX
    Explanations

    The neuron consistently activates on content‐heavy nouns and named entities—i.e. salient keywords like proper names, specific objects, or other key referential terms.

    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
    드는
    0.88
     తా
    0.85
    빽
    0.84
    가는
    0.84
     missão
    0.84
    VCTarget
    0.83
     går
    0.82
     ကျွန်
    0.82
    지는
    0.82
    ﻨ
    0.80
    POSITIVE LOGITS
    اوم
    0.78
    叟
    0.75
    ué
    0.73
    maxWidth
    0.72
    ات
    0.72
    tag
    0.72
    ao
    0.69
    噻
    0.69
    apeutics
    0.68
    cated
    0.67
    Activations Density 0.001%

    No Known Activations