© 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-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 57481
    Prev
    Next
    INDEX
    Explanations

    `is` `override` `library` `example`

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

    The neuron fires on the last word or particle at the end of a clause or sentence (i.e. the clause‐final token just before punctuation).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     réservation
    -1.88
    鋨
    -1.86
    the
    -1.84
    Quite
    -1.81
    Also
    -1.74
    Just
    -1.73
    Like
    -1.66
     compteur
    -1.64
    Some
    -1.63
    Take
    -1.63
    POSITIVE LOGITS
     Of
    2.27
    当然
    1.82
    zość
    1.70
     Then
    1.69
     poza
    1.58
     s
    1.57
    })=\
    1.55
    哎呀
    1.54
     generic
    1.52
     ‎
    1.52
    Activations Density 0.001%

    No Known Activations