© 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. 116601
    Prev
    Next
    INDEX
    Explanations

    difficult times

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

    The neuron strongly activates on very high‐frequency common English function words (e.g. articles, pronouns, prepositions).

    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
     esk
    -0.85
     studenten
    -0.82
    krieg
    -0.79
     sportivo
    -0.79
    crash
    -0.79
    ¡¡¡
    -0.79
     ristorante
    -0.78
     современных
    -0.77
     poë
    -0.76
     prachtig
    -0.76
    POSITIVE LOGITS
     difficult
    2.33
     challenging
    1.60
    Difficult
    1.59
     tough
    1.58
     times
    1.54
    difficult
    1.53
     turbulent
    1.52
     Difficult
    1.51
     time
    1.44
     troubled
    1.44
    Activations Density 0.027%

    No Known Activations