© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 73596
    Prev
    Next
    INDEX
    Explanations

    separating or separated items

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

    The neuron fires on occurrences of the word “separate” (and its morphological variants like “separating” or “separation”).

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
    .
    -2.45
    蠵
    -2.38
     The
    -2.30
     stava
    -2.23
    en
    -2.20
     '
    -2.16
     advent
    -2.13
     Now
    -2.08
     
    -2.05
    но
    -2.02
    POSITIVE LOGITS
    ୌ
    3.22
    そういえば
    2.56
    ↀ
    2.52
    ☫
    2.52
    玧
    2.52
    辚
    2.45
    gencias
    2.42
    おわりに
    2.41
    .’
    2.38
    ftrag
    2.38
    Activations Density 0.021%

    No Known Activations