© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 31681
    Prev
    Next
    INDEX
    Explanations

    achieve success in careers

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

    The neuron fires on marketing‐style, second-person promotional language—calls to action, persuasive headlines, and “if you want…you’ll…” ad copy.

    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
    GRAF
    -1.00
     things
    -0.98
    件事情
    -0.98
    DEFINITIONS
    -0.97
    ínű
    -0.94
     buen
    -0.91
     некото
    -0.90
     возник
    -0.89
    掘り
    -0.89
     if
    -0.88
    POSITIVE LOGITS
     endlich
    1.27
     sofort
    1.15
    olusi
    1.09
     transform
    1.09
     pratiquement
    1.00
     BOTH
    0.99
     stratég
    0.98
    WITHOUT
    0.98
     automatiquement
    0.96
    arming
    0.96
    Activations Density 0.090%

    No Known Activations