© 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. 10-GEMMASCOPE-RES-131K
    4. 39169
    Prev
    Next
    INDEX
    Explanations

    iron ore processing

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

    The neuron fires on occurrences of the substring “ore” (as in “ore” or “ores”)—i.e. mineral‐ore mentions.

    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
    ’
    -3.20
    ma
    -2.55
    But
    -2.55
    ه
    -2.55
    臯
    -2.45
    ks
    -2.45
    ”،
    -2.41
    ");
    -2.38
    ate
    -2.38
    hi
    -2.38
    POSITIVE LOGITS
     รวม
    2.59
    
    2.48
     cristianos
    2.33
     ciento
    2.28
    lardan
    2.25
     they
    2.23
    larda
    2.20
     ore
    2.20
    tır
    2.17
    mları
    2.16
    Activations Density 0.004%

    No Known Activations