© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 73732
    Prev
    Next
    INDEX
    Explanations

    early stage or life event

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

    The neuron detects the word "Early" at the beginning of paragraphs or sections in academic and informational texts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron activates specifically on the word “Early,” marking initial or beginning stages in a text.

    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
    ↵
    -3.69
      
    -3.23
     và
    -2.67
    倏
    -2.56
     “
    -2.47
    迎来
    -2.45
    十分的
    -2.44
     considerando
    -2.38
     présenter
    -2.34
    紧接着
    -2.34
    POSITIVE LOGITS
     शुर
    2.95
    ilares
    2.83
    メリカ
    2.78
     epitom
    2.59
    骈
    2.45
    t
    2.39
    咨
    2.38
    what
    2.38
    b
    2.38
     plethora
    2.36
    Activations Density 0.030%

    No Known Activations