© 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. 22-GEMMASCOPE-RES-131K
    7. 27264
    Prev
    Next
    INDEX
    Explanations

    talk about

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

    The neuron fires on occurrences of “talk” (in various forms like “Talk,” “talked,” “talking,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    ↵
    -3.47
     \
    -3.42
     –
    -3.05
    s
    -2.30
    </b>
    -2.28
     —
    -2.13
     ''
    -2.09
    诙
    -2.08
    1
    -2.08
      
    -2.05
    POSITIVE LOGITS
     about
    2.83
    هو
    2.47
    筚
    2.36
     夜景
    2.27
    琤
    2.20
    
    2.19
    ()",
    2.17
    er
    2.17
     苹果
    2.16
    𝒟
    2.16
    Activations Density 0.014%

    No Known Activations