© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 65290
    Prev
    Next
    INDEX
    Explanations

    carbon chemistry and bonds

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

    chemical structures and bonds, particularly carbon-based molecular components in scientific articles.

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

    The neuron specializes in spotting inline chemical formula and bond‐notation tokens (e.g. elemental symbols like “C” or “H,” bond dashes, stereochemical descriptors) in the text.

    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
     numberWith
    -0.70
    temun
    -0.69
     calcula
    -0.69
    Constitu
    -0.69
     besucht
    -0.69
     Warwickshire
    -0.68
    iint
    -0.68
     getEmail
    -0.67
     čin
    -0.67
     Wizards
    -0.66
    POSITIVE LOGITS
     carbon
    2.06
    carbon
    1.69
     carbons
    1.55
    Carbon
    1.54
     Carbon
    1.49
     carbono
    1.44
    碳
    1.39
     CARBON
    1.33
     diamond
    1.12
     saturated
    1.12
    Activations Density 0.092%

    No Known Activations