© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 4939
    Prev
    Next
    INDEX
    Explanations

    self

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

    This neuron detects the word “self,” especially when it appears as part of a compound (e.g. self-service, self-discovery, self-inflicted).

    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
     damos
    0.66
     tombol
    0.62
     pants
    0.59
     jackets
    0.58
     mustang
    0.58
    骓
    0.57
     influenza
    0.55
     tomar
    0.55
     televisión
    0.55
    လိုက်
    0.55
    POSITIVE LOGITS
     Self
    1.45
    Self
    1.41
     self
    1.38
     सेल्फ
    1.34
     SELF
    1.25
    self
    1.12
    自
    1.10
     само
    1.09
     自
    1.00
     Selbst
    1.00
    Activations Density 0.093%

    No Known Activations