© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 63363
    Prev
    Next
    INDEX
    Explanations

    starts phrases with specific words

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

    This neuron fires on programming‐language identifiers, especially class and type names in PascalCase.

    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
    JsonFormat
    -0.93
    cisme
    -0.90
    ἑ
    -0.89
    宝物
    -0.82
    íř
    -0.81
    Unmarshal
    -0.77
    JoinColumn
    -0.77
    JsonKey
    -0.74
     Goods
    -0.72
    hata
    -0.70
    POSITIVE LOGITS
     durability
    1.05
     wearer
    1.03
     charges
    1.02
     properties
    0.90
     usage
    0.90
     worn
    0.88
     embedded
    0.88
    charges
    0.86
     wear
    0.84
    壊
    0.83
    Activations Density 0.016%

    No Known Activations