© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 55603
    Prev
    Next
    INDEX
    Explanations

    derivatives, salts, portions thereof

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

    scientific/technical document elements like chemical compounds, methods, and parts of systems.

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

    The neuron flags the formal, multi-syllable technical nouns you see in patent claims—words naming components or processes (e.g. “derivatives,” “manufacturing,” “compositions,” “portions,” “salts,” etc.).

    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
     darin
    -1.02
    徬
    -0.90
    为此
    -0.84
    从中
    -0.84
    在这个
    -0.79
    ందు
    -0.78
     therein
    -0.78
     Possibilities
    -0.77
    竄
    -0.77
    在這
    -0.77
    POSITIVE LOGITS
     thereof
    2.75
     שלו
    0.91
     yours
    0.81
     davon
    0.77
     screaming
    0.75
     שלה
    0.75
    مول
    0.75
    ˟
    0.75
    EXC
    0.75
     manifiesto
    0.74
    Activations Density 0.051%

    No Known Activations