© 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. 34-GEMMASCOPE-RES-131K
    7. 2051
    Prev
    Next
    INDEX
    Explanations

    health, science, and social terms

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

    sentence or paragraph breaks and formatting structure in text.

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

    The neuron flags unusually rare or distinctive tokens—things like exclamation marks, standalone numbers, and infrequent technical or domain-specific words.

    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
    uteurs
    -1.04
    ilta
    -0.98
    -0.96
    😶
    -0.94
    formatie
    -0.94
    …"
    -0.94
    😐
    -0.94
    TBA
    -0.92
    😑
    -0.88
     nggak
    -0.88
    POSITIVE LOGITS
     couldnt
    1.57
     its
    1.45
     wouldnt
    1.39
    Its
    1.38
     Ive
    1.31
    Ive
    1.28
    Dont
    1.24
     dont
    1.24
     didnt
    1.20
     isnt
    1.20
    Activations Density 0.000%

    No Known Activations