© 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. 4180
    Prev
    Next
    INDEX
    Explanations

    states and domains

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

    This neuron activates on semantically meaningful content words—especially nouns and verbs—rather than on common 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
    áról
    -0.80
     ilyen
    -0.79
     як
    -0.79
    𝘼
    -0.77
     įva
    -0.75
    tudi
    -0.73
    vins
    -0.72
    зм
    -0.71
    idem
    -0.71
     pady
    -0.70
    POSITIVE LOGITS
     behaviors
    1.70
     practices
    1.70
     things
    1.67
     olyan
    1.67
     situations
    1.63
     qualities
    1.58
     conditions
    1.54
     something
    1.52
     types
    1.52
     levels
    1.51
    Activations Density 0.197%

    No Known Activations