© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 93060
    Prev
    Next
    INDEX
    Explanations

    way of doing something

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

    The neuron lights up on semantically rich “content” words—nouns, verbs, adjectives, and adverbs—rather than on purely functional or grammatical tokens.

    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
     tigre
    -1.56
     remedio
    -1.50
     ingrediente
    -1.50
     extenso
    -1.45
    ！』
    -1.39
     dialami
    -1.34
    //*/
    -1.34
     Templo
    -1.33
     artikkelen
    -1.33
     diinginkan
    -1.30
    POSITIVE LOGITS
     jedną
    1.98
    4
    1.69
    '
    1.66
    8
    1.60
    9
    1.53
    3
    1.52
    為に
    1.49
    7
    1.49
     alcune
    1.48
    "
    1.48
    Activations Density 0.008%

    No Known Activations