© 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. 34-GEMMASCOPE-RES-131K
    7. 103963
    Prev
    Next
    INDEX
    Explanations

    news article titles

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

    The neuron strongly activates on programming‐style identifiers and code tokens (e.g. class, method, field names with underscores and digits, file‐path components, and other alphanumeric code fragments).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     ২০১
    -1.80
     ۲۰۱
    -1.61
     Skype
    -1.61
     Ebola
    -1.56
     http
    -1.49
    Coronavirus
    -1.46
     Trump
    -1.38
     tslint
    -1.37
     Zika
    -1.35
    coronavirus
    -1.32
    POSITIVE LOGITS
     ChatGPT
    2.52
     metaverse
    2.28
     NFTs
    2.25
     TikTok
    1.96
     Metaverse
    1.88
     NFT
    1.85
     Wordle
    1.81
     Türkiye
    1.78
     Omicron
    1.70
    NFTs
    1.70
    Activations Density 0.482%

    No Known Activations