© 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. 9437
    Prev
    Next
    INDEX
    Explanations

    advantage, case, against

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

    The neuron detects words that signal advantage or favor—terms like “advantage,” “tilt,” “balance,” or “favour.”

    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
    łym
    -1.00
     गरे
    -0.98
    thun
    -0.85
     rift
    -0.84
    =?,
    -0.82
     retrovisor
    -0.81
     estratégico
    -0.80
     legjobb
    -0.79
    читан
    -0.79
     Assy
    -0.79
    POSITIVE LOGITS
     against
    1.17
    against
    1.16
     factors
    1.16
     factor
    1.13
    因素
    1.03
    今回は
    1.00
     credibility
    0.99
     Factors
    0.96
    ingly
    0.95
    ‌.
    0.93
    Activations Density 0.032%

    No Known Activations