© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 83277
    Prev
    Next
    INDEX
    Explanations

    exactly confirming what was stated

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

    This neuron detects emphatic confirmation phrases, such as “that’s” and “exactly,” used to assert or underscore a statement.

    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
    equal
    -0.81
     muhimu
    -0.80
    either
    -0.79
     thôi
    -0.78
    otherwise
    -0.76
    あって
    -0.75
     Either
    -0.74
    particular
    -0.73
     tampoco
    -0.73
    Even
    -0.71
    POSITIVE LOGITS
     exactly
    5.75
    Exactly
    4.16
    exactly
    4.09
     precisely
    4.00
     Exactly
    3.75
     exactamente
    3.58
     genau
    3.45
     exactement
    3.44
     exatamente
    3.25
     esattamente
    3.02
    Activations Density 0.085%

    No Known Activations