© 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. 34-GEMMASCOPE-RES-131K
    7. 87096
    Prev
    Next
    INDEX
    Explanations

    off

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

    This neuron flags content-bearing (non–function) words—i.e. rarer nouns, verbs, and adjectives rather than common stopwords.

    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
    òm
    -0.82
    ϯ
    -0.80
    stus
    -0.79
    zyp
    -0.74
    ével
    -0.73
    આ
    -0.73
    onViewCreated
    -0.73
    тиза
    -0.71
    앓
    -0.71
     Estad
    -0.71
    POSITIVE LOGITS
     off
    8.31
    Off
    4.69
    off
    4.69
     Off
    4.16
     OFF
    3.78
    OFF
    3.28
    オフ
    2.95
     ऑफ
    2.41
    offs
    2.39
     offs
    2.20
    Activations Density 0.199%

    No Known Activations