© 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. 10-GEMMASCOPE-RES-131K
    7. 43822
    Prev
    Next
    INDEX
    Explanations

    interesting followed by detail

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

    The neuron flags evaluative adjectives and nouns indicating that something is of special interest or importance in the text.

    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
    before
    -1.84
    with
    -1.53
          
    -1.50
    ERS
    -1.48
     Parteien
    -1.44
     entweder
    -1.39
    Either
    -1.38
     laiko
    -1.38
    
    -1.35
        
    -1.33
    POSITIVE LOGITS
    </em>
    1.30
     even
    1.28
     because
    1.27
     сия
    1.27
     ourselves
    1.27
     ग्राहकों
    1.26
     ilyen
    1.24
    硤
    1.23
    psie
    1.22
    suo
    1.21
    Activations Density 0.052%

    No Known Activations