© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 1479
    Prev
    Next
    INDEX
    Explanations

    difficulties and challenges

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

    The main thing this neuron does is detect adjectives and descriptive words that convey difficulty, stress, or being overwhelmed.

    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
     that
    -1.28
    pergillus
    -0.93
     likely
    -0.91
     either
    -0.90
    esinde
    -0.89
     unlikely
    -0.88
     viață
    -0.88
     protegida
    -0.88
     quintessential
    -0.88
     exceptional
    -0.87
    POSITIVE LOGITS
     especially
    1.61
    Especially
    1.42
     at
    1.41
     особенно
    1.34
     особли
    1.26
     especialmente
    1.20
    especially
    1.19
     Особенно
    1.19
    尤其是
    1.19
     trying
    1.18
    Activations Density 0.042%

    No Known Activations