© 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. 31-GEMMASCOPE-2-RES-16K
    4. 92
    Prev
    Next
    INDEX
    Explanations

    link to projects or store

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

    The neuron fires strongly on content‐carry­ing words (longer, semantically rich tokens) rather than on common short function words.

    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_31_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
     
    0.44
    etheless
    0.44
     resonate
    0.42
    9
    0.42
     आस्ट्रेलिया
    0.41
    0
    0.40
     nonetheless
    0.40
     nonstop
    0.40
    2
    0.40
     sculpted
    0.40
    POSITIVE LOGITS
     Фор
    0.54
     comunidad
    0.49
    <unused631>
    0.48
    <unused581>
    0.47
     sostuvo
    0.47
    <unused1884>
    0.46
    Ⴀ
    0.46
     sogenannten
    0.46
    <unused1016>
    0.45
     Бан
    0.45
    Activations Density 0.000%

    No Known Activations