© 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. 10-GEMMASCOPE-RES-131K
    7. 20447
    Prev
    Next
    INDEX
    Explanations

    technical documentation snippets

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

    The neuron is specialized to pick out numeric and separator tokens (e.g. line numbers, hyphens) used in code‐diff or patch metadata (hunk headers, file indices).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    ],
    -1.42
    気に入りの
    -1.40
     мені
    -1.37
     interessiert
    -1.34
     beschäftigt
    -1.34
     beschäf
    -1.33
     fordert
    -1.30
    そして
    -1.30
     •
    -1.30
     със
    -1.27
    POSITIVE LOGITS
    Ǥ
    1.42
    瓤
    1.36
     masih
    1.35
     unica
    1.32
     comprende
    1.30
     libera
    1.29
     spazi
    1.29
    一层
    1.29
    ‘’
    1.29
     tinte
    1.28
    Activations Density 0.004%

    No Known Activations