© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 62627
    Prev
    Next
    INDEX
    Explanations

    states and qualities

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

    The neuron strongly activates on descriptive adjectives that characterize or qualify nouns (i.e. property‐ or quality-words).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
     its
    -2.05
     it
    -1.84
     itself
    -1.54
    它
    -1.36
     was
    -1.30
     is
    -1.22
    它是
    -1.17
    it
    -1.16
    h
    -1.09
    its
    -1.08
    POSITIVE LOGITS
     themselves
    2.48
    themselves
    1.98
     their
    1.59
     entities
    1.55
     jejich
    1.52
     έχουν
    1.52
     creatures
    1.50
     eivät
    1.46
     olduk
    1.46
     beasts
    1.43
    Activations Density 0.244%

    No Known Activations