© 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. 84854
    Prev
    Next
    INDEX
    Explanations

    sourdough starter bread

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

    formatting elements in documents such as punctuation, special characters, and layout markers.

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

    This neuron fires strongly on recipe-style ingredient listings—words and tokens that denote quantities, food items, and measurement units in a cooking recipe.

    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
     nutmeg
    -0.86
     Hebrews
    -0.75
     Binghamton
    -0.73
     outcrop
    -0.71
     ornith
    -0.70
     daff
    -0.69
    teira
    -0.69
    Spinach
    -0.68
     breeze
    -0.68
    birthDate
    -0.68
    POSITIVE LOGITS
     starter
    1.21
     hydration
    1.19
    hydration
    1.12
     método
    0.97
     starters
    0.93
     scoring
    0.93
     sourdough
    0.88
     Starter
    0.87
    starter
    0.85
     leva
    0.85
    Activations Density 0.005%

    No Known Activations