© 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. 34-GEMMASCOPE-RES-131K
    4. 25652
    Prev
    Next
    INDEX
    Explanations

    survival activities

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

    This neuron detects function words that mark second‐person hypothetical questions—e.g., “you,” “would,” “what,” and “if you” in question contexts.

    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
     floodplain
    -0.88
    对手
    -0.87
    ulica
    -0.84
    їн
    -0.82
     pauper
    -0.81
    digung
    -0.81
    mafia
    -0.80
     larceny
    -0.80
    censo
    -0.79
    bors
    -0.78
    POSITIVE LOGITS
     survivors
    2.31
     survivor
    2.06
     survival
    1.91
    Survivors
    1.71
     Survivors
    1.66
    survi
    1.66
     surviving
    1.65
    survival
    1.41
    Survival
    1.38
     supplies
    1.37
    Activations Density 0.015%

    No Known Activations