© 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. 22-GEMMASCOPE-RES-131K
    7. 79848
    Prev
    Next
    INDEX
    Explanations

    learning new skills and knowledge

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

    The neuron detects instructional or educational language—especially words and phrases that describe teaching, training, or acquiring skills (e.g. “learn,” “training,” “how to,” “using”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    licante
    -1.18
    实在是
    -1.11
     simplemente
    -1.11
     съдържа
    -1.08
     этими
    -1.07
     afbeelding
    -1.05
    只是一个
    -1.04
    してしまう
    -1.03
    prehen
    -1.02
     característica
    -1.02
    POSITIVE LOGITS
     how
    2.13
     techniques
    1.78
     learn
    1.66
     skills
    1.55
     understand
    1.48
     knowledge
    1.42
     basic
    1.41
     concepts
    1.40
     basics
    1.39
     topics
    1.38
    Activations Density 0.047%

    No Known Activations