© 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
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 5077
    Prev
    Next
    INDEX
    Explanations

    The neuron is essentially acting like a “content‐word” detector: it fires on semantically informative words (nouns, verbs, adjectives, numbers, etc.) rather than on common function words.

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     &
    0.89
     (<
    0.81
     (&
    0.80
     (%
    0.78
     (~
    0.77
    /
    0.75
     waxay
    0.74
     &/
    0.73
     (
    0.72
     (>
    0.72
    POSITIVE LOGITS
    。,
    1.03
    。【
    1.00
    。.
    0.98
     ।,
    0.82
    ؟.
    0.80
    。...
    0.79
     ');
    0.79
     };
    0.77
    .""
    0.77
    ."-
    0.77
    Activations Density 0.819%

    No Known Activations