© 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. 34-GEMMASCOPE-RES-131K
    7. 31357
    Prev
    Next
    INDEX
    Explanations

    understanding or recognizing

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

    The neuron activates most strongly on the key action words (verbs and auxiliaries) in instructional or explanatory sentences—i.e. the core verbs and helping verbs that carry the “advice/explanation” tone.

    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
    ánicos
    -0.98
    Comparer
    -0.85
    філь
    -0.84
    nachweise
    -0.83
     ($('#
    -0.82
    løs
    -0.81
    fw
    -0.80
    CPtr
    -0.79
     colspan
    -0.78
     foc
    -0.78
    POSITIVE LOGITS
     read
    2.55
     reading
    2.30
     sensing
    2.16
     reads
    2.06
     intu
    2.05
     sensed
    1.99
     guessed
    1.81
     detect
    1.77
    read
    1.71
     detects
    1.67
    Activations Density 0.030%

    No Known Activations