© 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. 22-GEMMASCOPE-RES-131K
    7. 39385
    Prev
    Next
    INDEX
    Explanations

    duplicate entry

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

    This neuron activates on technical entity names—nouns referring to data or code objects (e.g. journal entries, folders, commissions, user IDs, records, model classes).

    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
    ただし
    -0.82
    Ventajas
    -0.82
     sendMessage
    -0.81
     extra
    -0.77
    éfonos
    -0.77
     filtered
    -0.77
    禿
    -0.75
     طبق
    -0.75
    Damon
    -0.74
     bulunur
    -0.74
    POSITIVE LOGITS
     similar
    1.17
     same
    1.06
     identical
    1.02
     podob
    0.96
    同じ
    0.94
     сход
    0.92
    ähn
    0.91
    Same
    0.91
     serupa
    0.91
     benzer
    0.90
    Activations Density 0.016%

    No Known Activations