© 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. 10-GEMMASCOPE-RES-131K
    7. 69280
    Prev
    Next
    INDEX
    Explanations

    elements like boron and BH3

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

    chemical and mathematical symbols, particularly chemical elements and variables in formulas.

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

    The neuron fires primarily on numeric quantity tokens (e.g. measured values and element counts).

    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
    荭
    -2.63
    acterium
    -2.25
     —
    -2.25
    ᕱ
    -2.20
    哐
    -2.20
    ⸎
    -2.09
     くれ
    -2.08
    ケティング
    -2.08
    咣
    -2.03
    ‐
    -2.00
    POSITIVE LOGITS
    2
    3.94
    3
    3.45
    9
    3.00
    8
    2.97
    This
    2.97
    7
    2.92
    From
    2.88
    6
    2.73
    before
    2.70
    with
    2.64
    Activations Density 5.882%

    No Known Activations