© 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. 22-GEMMASCOPE-RES-131K
    7. 16502
    Prev
    Next
    INDEX
    Explanations

    dataset, elimination, machine learning

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

    The neuron activates on programming code tokens—especially identifiers like variable and function names (and their adjacent symbols) in code snippets.

    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_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     gunakan
    -1.22
    UnityEngine
    -1.20
     keramik
    -1.17
     дѣ
    -1.14
     OBITUARY
    -1.14
    мѣ
    -1.09
     keny
    -1.08
     acredita
    -1.08
     Obispo
    -1.08
     maleta
    -1.07
    POSITIVE LOGITS
    Recept
    1.10
    というわけで
    1.08
     but
    1.07
    iencies
    1.05
     đứa
    1.01
    以上です
    1.00
     nên
    0.99
     олив
    0.98
    ONESIA
    0.98
     насеље
    0.98
    Activations Density 0.001%

    No Known Activations