© 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. 10-GEMMASCOPE-RES-131K
    7. 45135
    Prev
    Next
    INDEX
    Explanations

    Molly or Dirac paired with names or physics terms

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

    The neuron fires on domain‐specific technical identifiers (e.g. class names or specialized scientific terms such as “Dirac,” “HttpResponse,” or “tetra”)

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    
    -3.63
    沨
    -3.16
     gustaMe
    -3.05
    锝
    -2.89
    commonly
    -2.83
     feront
    -2.83
    鶿
    -2.77
     spearheaded
    -2.75
     raiſ
    -2.73
    翃
    -2.72
    POSITIVE LOGITS
    When
    3.75
    Despite
    3.22
    While
    3.16
    Even
    3.08
    After
    2.98
    It
    2.95
    One
    2.94
    Before
    2.91
    Just
    2.89
    During
    2.89
    Activations Density 0.007%

    No Known Activations