© 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
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 10516
    Prev
    Next
    INDEX
    Explanations

    his own possession

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

    The neuron responds to semantically salient content tokens—especially proper nouns, pronouns, quantifiers, and emphasis words—in expository or persuasive passages.

    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_31_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
     
    0.72
    ‏
    0.71
    \/
    0.70
    /=
    0.67
    (
    0.57
     Rum
    0.56
    Checkpoint
    0.55
    ↵
    0.55
    \
    0.55
    	
    0.55
    POSITIVE LOGITS
     because
    1.66
     during
    1.57
     puisqu
    1.57
    !”,
    1.52
     when
    1.48
     eftersom
    1.45
     until
    1.41
     όταν
    1.41
     çünkü
    1.41
     puisque
    1.41
    Activations Density 1.953%

    No Known Activations