© 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. 10-GEMMASCOPE-RES-131K
    7. 90808
    Prev
    Next
    INDEX
    Explanations

    recent past actions

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

    The neuron fires on words that signal recent or past actions—news‐style update markers like “recently,” past‐tense verbs (“did,” “launched,” “reported,” etc.), and similar temporal cues.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     opulent
    -1.70
    Eventually
    -1.61
    paramInt
    -1.59
     verschieden
    -1.58
     horrid
    -1.58
     construit
    -1.56
    跛
    -1.52
    Either
    -1.49
     créée
    -1.46
     wacky
    -1.45
    POSITIVE LOGITS
     two
    2.08
     because
    2.05
     three
    1.82
     helps
    1.76
     their
    1.74
     cleverly
    1.69
     although
    1.69
     superbly
    1.66
     makes
    1.64
     helped
    1.63
    Activations Density 0.046%

    No Known Activations