© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 57024
    Prev
    Next
    INDEX
    Explanations

    approximations of quantity

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

    The neuron activates strongly on words that signal approximations of quantity or time (e.g. “just,” “close,” “somewhere”)—i.e. vague or approximate numeric qualifiers.

    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
     años
    -1.13
    しかも
    -0.98
     tragé
    -0.93
     especially
    -0.83
    astră
    -0.83
    ſen
    -0.82
     даже
    -0.81
    ząt
    -0.80
     besonders
    -0.80
     brisa
    -0.80
    POSITIVE LOGITS
     حدود
    1.03
    выше
    0.96
    nearly
    0.96
    めました
    0.93
    nigh
    0.93
     whopping
    0.91
    為に
    0.91
     neigh
    0.88
     approaching
    0.84
     shy
    0.83
    Activations Density 0.031%

    No Known Activations