© 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. 93431
    Prev
    Next
    INDEX
    Explanations

    expressions of negation or denial

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

    The neuron picks out words and short phrases that signal strong subjective emphasis or hedging of an opinion (e.g. “obsessed,” “at all,” “at least,” “claims,” “know,” “see”).

    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
    itiés
    -0.88
    '&:
    -0.85
    douard
    -0.84
     unfailing
    -0.83
    avant
    -0.82
     liksom
    -0.79
    BURGH
    -0.78
     Algunas
    -0.77
    uza
    -0.77
     <<<<<<<<<<<<<<
    -0.76
    POSITIVE LOGITS
     any
    1.05
     nothing
    0.82
     dijeron
    0.82
    interest
    0.80
    Interest
    0.78
     никаких
    0.78
    有任何
    0.78
    écart
    0.77
     сър
    0.77
    
    0.75
    Activations Density 0.036%

    No Known Activations