© 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. 38673
    Prev
    Next
    INDEX
    Explanations

    Instagram handles

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

    This neuron detects social‐media metadata tokens, especially user handles (e.g. “@username”) and their associated numeric counts.

    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
     twitter
    -1.13
     tweet
    -1.08
     tweeted
    -0.99
     Twitter
    -0.99
    twitter
    -0.94
     ottobre
    -0.91
     would
    -0.89
     lutego
    -0.89
     stycznia
    -0.89
    tweet
    -0.89
    POSITIVE LOGITS
    ._
    1.60
     official
    1.44
    official
    1.42
     Instagram
    1.33
    ._.
    1.32
    ⠀⠀⠀⠀⠀⠀⠀⠀
    1.28
    Official
    1.26
    _.
    1.25
     Official
    1.18
    .__
    1.17
    Activations Density 0.003%

    No Known Activations