© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 93582
    Prev
    Next
    INDEX
    Explanations

    raising awareness

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

    The neuron is highly activated by occurrences of “awareness” (and related forms), effectively detecting mentions of raising or highlighting awareness.

    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
     ‘
    -0.92
     references
    -0.87
     colder
    -0.84
     philosophy
    -0.81
     urge
    -0.80
     v
    -0.79
    ритори
    -0.79
    li
    -0.79
     "
    -0.79
     overweight
    -0.79
    POSITIVE LOGITS
     raising
    1.91
    raising
    1.76
     Raising
    1.64
     raise
    1.59
    awareness
    1.46
    raise
    1.46
    Raising
    1.43
     Raise
    1.41
     awareness
    1.38
     raised
    1.37
    Activations Density 0.012%

    No Known Activations