© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 76672
    Prev
    Next
    INDEX
    Explanations

    short-handed, injury, depleted

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

    The neuron fires on the punctuation-heavy, “newswire” style clauses and parenthetical details (times, commas, parentheses, dateline-type qualifiers) you see in journalistic sentences.

    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
     carrière
    -0.82
     FAILURE
    -0.81
     торже
    -0.80
     religieuses
    -0.79
    Eloquent
    -0.79
    ℒ
    -0.76
     almendras
    -0.75
    かす
    -0.75
     социали
    -0.73
    Failure
    -0.73
    POSITIVE LOGITS
     depleted
    1.84
     injuries
    1.64
     injury
    1.41
     absences
    1.41
     deple
    1.39
    Injuries
    1.38
     missing
    1.38
     depletion
    1.33
     decim
    1.27
     depth
    1.21
    Activations Density 0.034%

    No Known Activations