© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 54377
    Prev
    Next
    INDEX
    Explanations

    chemical formulas and names

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

    chemical formulas and molecular compound notations.

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

    The neuron spots the “~” characters (and the numbers they connect to) used to denote approximate element counts in chemical formula/crystal‐data lines.

    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
     which
    -1.22
     or
    -1.09
     even
    -1.05
    אָ
    -1.01
     💀
    -0.99
    hhhhhhhh
    -0.98
     when
    -0.97
    ఘ
    -0.96
    руб
    -0.95
    运算
    -0.94
    POSITIVE LOGITS
     trois
    1.43
     deux
    1.22
     作り方
    1.18
     conster
    1.16
     quatre
    1.16
    푉
    1.13
    そうですが
    1.09
    Série
    1.07
    loys
    1.06
    ):
    1.05
    Activations Density 0.006%

    No Known Activations