© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 6865
    Prev
    Next
    INDEX
    Explanations

    recommendations or suggestions

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

    The main thing this neuron does is spot strong recommendation phrases (e.g. “recommend,” “would definitely stay again,” “highly recommend”).

    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_31_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
     devastation
    0.59
     Return
    0.58
    ReturnValue
    0.58
    愿
    0.57
    Return
    0.56
     will
    0.55
     Gods
    0.55
     impec
    0.55
    願
    0.55
     return
    0.54
    POSITIVE LOGITS
     anyone
    1.11
    anyone
    1.08
     fortement
    1.02
     anybody
    1.02
     iedereen
    0.99
    給大家
    0.97
    任何人
    0.95
     strongly
    0.93
    Anybody
    0.92
     Anyone
    0.91
    Activations Density 0.095%

    No Known Activations