© 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. 34-GEMMASCOPE-RES-131K
    7. 91478
    Prev
    Next
    INDEX
    Explanations

    other

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

    The neuron is primarily detecting ordinary content words (open‐class vocabulary like nouns, verbs, and adjectives) rather than punctuation or purely grammatical function words.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    雖然
    -1.14
    ##
    -1.10
    ticularly
    -1.05
    specifically
    -1.04
     zowel
    -1.04
    particularly
    -1.00
    久しぶりに
    -1.00
     nejen
    -0.98
    虽然
    -0.98
     entrambi
    -0.98
    POSITIVE LOGITS
     other
    2.44
     других
    1.85
     analogous
    1.70
     otros
    1.63
    其他的
    1.56
    Other
    1.54
    其他
    1.51
     Other
    1.51
    其它
    1.50
     elsewhere
    1.49
    Activations Density 0.250%

    No Known Activations