© 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. 84430
    Prev
    Next
    INDEX
    Explanations

    hard work and difficulty

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

    The neuron activates on occurrences of the word “hard” (including compounds or inflected forms beginning with “hard-”).

    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
    MediaType
    -0.82
    Zab
    -0.80
    Gecko
    -0.79
     évêque
    -0.77
    Bety
    -0.77
    꿨
    -0.75
     familiares
    -0.75
     víc
    -0.74
     FLAGS
    -0.73
    vább
    -0.72
    POSITIVE LOGITS
     hard
    1.29
     charging
    1.19
     fought
    1.09
    Hard
    1.07
    charging
    0.91
    nosed
    0.90
    hard
    0.88
     HARD
    0.88
     Hard
    0.85
    HARD
    0.85
    Activations Density 0.031%

    No Known Activations