© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 9371
    Prev
    Next
    INDEX
    Explanations

    chemical formulas and notations

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

    The neuron activates on tokens representing precise scientific numerical values (especially decimal measurements such as mass/charge or molecular weight figures).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    alphanumeric codes and identifiers, particularly in scientific, technical, and catalog contexts.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    uppercase alphanumeric abbreviations and scientific shorthand, such as chemical formulas, biological/protein acronyms, subunit letters, and model or catalog codes.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     NodeJS
    0.52
     EBIT
    0.47
     palladium
    0.47
     नव्ह
    0.45
     crossword
    0.45
     pyridine
    0.44
     பொருந்த
    0.43
     usw
    0.43
     xpath
    0.43
     polymeric
    0.42
    POSITIVE LOGITS
    rifice
    0.47
    schluss
    0.46
     Tua
    0.43
     이유
    0.43
    র্
    0.43
    uable
    0.42
    ülle
    0.42
    ö
    0.42
    ertype
    0.42
    ijl
    0.41
    Activations Density 0.126%

    No Known Activations