© 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. 31-GEMMASCOPE-2-RES-16K
    7. 15972
    Prev
    Next
    INDEX
    Explanations

    Pig Latin `ay` words

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

    The neuron is picking out relatively rare, formal or “high-register” content words—especially sophisticated adjectives and nouns common in academic or technical writing.

    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
    初めて
    0.57
    renderCamera
    0.55
    ちょうど
    0.54
    クレジットカード
    0.52
    🏼
    0.52
    横浜
    0.52
    ニ
    0.51
    полне
    0.51
    ンダント
    0.51
    fromDate
    0.51
    POSITIVE LOGITS
     rely
    0.47
     stren
    0.46
     ofthe
    0.46
     offs
    0.46
     ampere
    0.46
     davon
    0.45
     byli
    0.45
     adenine
    0.45
     disso
    0.44
    -
    0.44
    Activations Density 0.197%

    No Known Activations