© 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. 28456
    Prev
    Next
    INDEX
    Explanations

    is, was, did, it

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

    The neuron fires most strongly on content‐bearing words—i.e. the key nouns and verbs carrying the main meaning of a sentence.

    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
    
    -1.02
    codile
    -1.02
     แต่
    -0.98
     אם
    -0.93
     infatti
    -0.93
    mediately
    -0.92
    fraid
    -0.91
    échal
    -0.91
     BECAUSE
    -0.89
    lavia
    -0.89
    POSITIVE LOGITS
     makes
    1.77
     is
    1.73
     was
    1.52
     I
    1.41
     it
    1.34
     just
    1.23
     really
    1.20
     definitely
    1.19
     certainly
    1.14
     all
    1.13
    Activations Density 0.018%

    No Known Activations