© 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. 104457
    Prev
    Next
    INDEX
    Explanations

    This is a class

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

    The neuron fires on non‐code words inside comments or docstrings (i.e. spans of explanatory or descriptive text in source comments).

    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
    <bos>
    -1.23
     아트
    -1.18
     last
    -1.17
    Karakter
    -1.16
     prachtige
    -1.13
     εκεί
    -1.11
     воспользова
    -1.10
    Lokal
    -1.09
     travailleurs
    -1.09
     그것
    -1.09
    POSITIVE LOGITS
     ajustable
    1.43
    pernicus
    1.30
     herman
    1.25
     Altos
    1.21
    ⓐ
    1.17
     MyBatis
    1.15
     smtplib
    1.13
    を行う
    1.12
     RadioButton
    1.12
     FANDOM
    1.11
    Activations Density 0.144%

    No Known Activations