© 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. 31054
    Prev
    Next
    INDEX
    Explanations

    rules and inference

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

    The neuron detects listings of formal logic inference or replacement rules (e.g. acronyms like MP, MT, HS, DS, Conj, etc.) and related rule‐listing syntax.

    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
     اختلاف
    -0.76
    眶
    -0.73
    intval
    -0.71
    DidChange
    -0.71
    Laurent
    -0.70
    🗣
    -0.70
     письмо
    -0.69
    lical
    -0.67
    nath
    -0.66
     Legendary
    -0.66
    POSITIVE LOGITS
     rule
    1.35
     rules
    1.27
    规则
    1.01
     Rule
    0.99
     правила
    0.97
     Wea
    0.96
     weakening
    0.95
    Rule
    0.91
     elimination
    0.91
     RULES
    0.90
    Activations Density 0.016%

    No Known Activations