© 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. 75919
    Prev
    Next
    INDEX
    Explanations

    farm

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

    This neuron detects standalone identifier-like tokens used as field or symbol names in structured data or code (e.g. metadata keys like “description”/“decimals,” package or module names, and other code identifiers).

    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
     ябло
    -0.75
    ⍵
    -0.69
     Nguyên
    -0.69
     przes
    -0.67
     solicitar
    -0.66
     Scor
    -0.66
    sandbox
    -0.66
    Scor
    -0.66
     prefeito
    -0.66
    饱
    -0.65
    POSITIVE LOGITS
     farms
    1.13
     farm
    1.06
     farming
    1.01
    Farm
    0.95
    farm
    0.94
     harvest
    0.94
     farmed
    0.93
    chef
    0.89
    farms
    0.89
     Farm
    0.88
    Activations Density 0.002%

    No Known Activations