© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 8475
    Prev
    Next
    INDEX
    Explanations

    NLP model names and technical terms

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

    The neuron flags terms and phrases typical of technical instructions or configuration documentation.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     socalled
    0.64
    
    0.63
     luscious
    0.60
    .
    0.60
    Caribbean
    0.57
    -
    0.57
    ​—
    0.56
    ­
    0.55
    ,’’
    0.54
     endowment
    0.53
    POSITIVE LOGITS
     github
    1.11
     onboarding
    1.08
     Github
    1.07
     workaround
    1.03
     GitHub
    1.01
     RMSE
    1.01
     overfitting
    1.01
     vscode
    1.00
     jquery
    0.98
     Laravel
    0.97
    Activations Density 0.092%

    No Known Activations