© 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. 22-GEMMASCOPE-RES-131K
    4. 12584
    Prev
    Next
    INDEX
    Explanations

    qualities and their conditions

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

    This neuron fires on descriptive (especially evaluative) adjectives and adverbs that qualify or comment on the complexity or quality of something.

    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
     but
    -2.61
     แต่
    -1.80
    ありますが
    -1.77
     nhưng
    -1.72
     ولكن
    -1.60
     tetapi
    -1.59
    但是
    -1.52
     αλλά
    -1.51
    かもしれませんが
    -1.50
     ngunit
    -1.45
    POSITIVE LOGITS
     it
    1.79
     if
    1.49
     because
    1.48
     there
    1.30
     still
    1.23
    </tr>
    1.19
     поскольку
    1.17
     becauſe
    1.15
     since
    1.13
    }
    1.09
    Activations Density 0.121%

    No Known Activations