© 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. 65711
    Prev
    Next
    INDEX
    Explanations

    past tense verbs

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

    The neuron detects the auxiliaries and adverbs that signal passive‐voice or official‐report style constructions (e.g. “were,” “was,” “promptly,” “stopped,” “initiated”).

    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
     Eröffnung
    -0.84
    ुत
    -0.82
     tunn
    -0.80
    ků
    -0.78
    افه
    -0.77
     Einladung
    -0.76
    discover
    -0.76
    tamp
    -0.74
    ARAN
    -0.74
    کیل
    -0.74
    POSITIVE LOGITS
    innis
    0.87
    methy
    0.83
    tadiene
    0.76
     commemorating
    0.75
    片
    0.75
    寝
    0.73
    dens
    0.73
    __(/*!
    0.73
     पर्
    0.72
    نامه
    0.72
    Activations Density 0.027%

    No Known Activations