© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 34927
    Prev
    Next
    INDEX
    Explanations

    purpose or intention

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

    The neuron fires on past‐tense verb forms (in particular irregular ones like “felt” and “meant”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    结语
    -2.88
    最后由
    -2.78
    勍
    -2.69
    鋮
    -2.56
    举例
    -2.45
    祧
    -2.45
    另一种
    -2.44
     decorada
    -2.42
    珖
    -2.41
    狆
    -2.39
    POSITIVE LOGITS
    2
    3.05
    "
    3.03
    1
    2.88
     inspiré
    2.58
     infused
    2.53
    需要
    2.52
    *
    2.47
    ’
    2.45
    \
    2.45
     fueled
    2.42
    Activations Density 0.041%

    No Known Activations