© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 7202
    Prev
    Next
    INDEX
    Explanations

    chemical-like structure delimiters

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

    chemical formulas and scientific notation in brackets or parentheses.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron is strongly activated by the closing bracket ‘]’ (especially in numbered or formulaic contexts), i.e. it detects the ends of bracketed expressions.

    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
    </sub>
    -2.09
     Kese
    -2.05
     '-';
    -2.02
     "")
    
    -1.94
     Gennaio
    -1.93
     raiſ
    -1.92
     Toscana
    -1.90
    賒
    -1.89
     ""),
    -1.88
     snart
    -1.85
    POSITIVE LOGITS
      
    2.25
     remporté
    2.06
    ]
    2.03
     Besonder
    2.02
    end
    1.96
    くださ
    1.95
    下さ
    1.89
     Of
    1.88
    珝
    1.86
    ...
    1.80
    Activations Density 0.097%

    No Known Activations