© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 14222
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on descriptive modifiers—especially adjectives (and adjectival past participles) that qualify or characterize entities.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    และความ
    0.80
     lightness
    0.79
     ৷
    0.75
     adaptability
    0.75
    ❕
    0.74
     darkness
    0.73
     который
    0.73
     digestibility
    0.72
     !”
    0.71
     и
    0.70
    POSITIVE LOGITS
    4
    0.85
    7
    0.85
    0
    0.84
    8
    0.82
    9
    0.81
    5
    0.80
    2
    0.77
    6
    0.77
    3
    0.75
    1
    0.72
    Activations Density 0.930%

    No Known Activations