© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 49183
    Prev
    Next
    INDEX
    Explanations

    later on

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

    This neuron detects discourse-level transition markers (words or short phrases like “however,” “therefore,” or “in this case”) that signal contrast or causal connection.

    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
     raha
    -0.86
    ultado
    -0.85
     hapa
    -0.82
     Coim
    -0.80
     huo
    -0.78
     story
    -0.77
     tombstone
    -0.74
    momento
    -0.73
    ʽ
    -0.73
     cr
    -0.73
    POSITIVE LOGITS
    Later
    0.85
    conda
    0.85
    0.84
     fotó
    0.84
     HFILL
    0.81
    0.81
     hyvin
    0.81
    culada
    0.79
    aks
    0.78
     LATER
    0.78
    Activations Density 0.007%

    No Known Activations