© 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. 19257
    Prev
    Next
    INDEX
    Explanations

    btw and interjections

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

    The neuron fires on sequences of (mostly short) all-caps letter strings—i.e. acronyms or initialisms (like “BTW,” “BBC,” “FFFF,” etc.).

    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
    '
    -3.36
    на
    -2.92
    ↵↵
    -2.81
     aumenta
    -2.47
     precisa
    -2.39
     
    -2.22
     normalmente
    -2.16
     principalmente
    -2.14
     of
    -2.14
            
    -2.14
    POSITIVE LOGITS
    媜
    2.84
     櫃
    2.67
    quelize
    2.55
    褔
    2.53
    
    2.48
     夜景
    2.48
     园
    2.45
    垻
    2.45
    ”،
    2.45
    豋
    2.41
    Activations Density 0.000%

    No Known Activations