© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 8119
    Prev
    Next
    INDEX
    Explanations

    verb-preposition phrases

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

    The neuron is detecting English verbs in their past-tense or past-participle form marked by the “-ed” suffix.

    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
    लिए
    -1.79
     precisamente
    -1.65
     россия
    -1.63
    それぞれ
    -1.59
    只需
    -1.59
     for
    -1.58
    鉛筆
    -1.55
    ла
    -1.54
    Balancer
    -1.54
    それぞれの
    -1.53
    POSITIVE LOGITS
     afferma
    1.73
     mattino
    1.69
    を始め
    1.66
    魠
    1.65
    “
    1.63
     nicio
    1.63
     zaczą
    1.63
     taglio
    1.63
    ↵↵↵
    1.60
     criou
    1.59
    Activations Density 0.069%

    No Known Activations