© 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. 90468
    Prev
    Next
    INDEX
    Explanations

    estimating cost and accuracy

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

    The neuron strongly activates on formal “term‐definition” passages—e.g. headings, acronyms in parentheses, and their explanatory sentences defining a concept.

    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
     bojo
    -0.81
    DateFormatter
    -0.78
    ketten
    -0.75
    ליון
    -0.74
     малю
    -0.73
    Дополнитель
    -0.72
    Console
    -0.72
     comprim
    -0.71
     rekomendasi
    -0.71
     ocio
    -0.70
    POSITIVE LOGITS
     estimating
    1.96
     estimate
    1.69
     estimator
    1.64
     Esti
    1.58
     Estimating
    1.45
     cost
    1.42
     estimation
    1.41
    Estimator
    1.41
     estimates
    1.38
     estimators
    1.38
    Activations Density 0.009%

    No Known Activations