© 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. 10-GEMMASCOPE-RES-131K
    4. 127279
    Prev
    Next
    INDEX
    Explanations

    inflating layouts or objects

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

    The neuron activates on tokens containing the “infl” root—e.g. inflate/Inflater/infal, inflated, inflatable.

    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
    It
    -2.63
    When
    -2.63
    They
    -2.61
    but
    -2.50
    These
    -2.42
    If
    -2.36
    可能
    -2.28
     交換
    -2.22
    According
    -2.19
    This
    -2.16
    POSITIVE LOGITS
    ⃢
    2.19
    jaqueta
    2.11
    ]
    2.00
    哙
    1.86
    umumkan
    1.86
     stray
    1.85
    mäs
    1.80
    踅
    1.80
    chaqueta
    1.76
     menti
    1.73
    Activations Density 0.002%

    No Known Activations