© 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. 7670
    Prev
    Next
    INDEX
    Explanations

    print("Invalid input")

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

    The neuron responds to phrases expressing reciprocal benefit or “give-and-take” notions (e.g. “give something back,” “get out of it,” “what’s in it for me”).

    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
    fasterxml
    0.60
     Coupling
    0.57
     Höhen
    0.56
     Pairs
    0.56
     Pretty
    0.53
    }?
    0.53
     Parts
    0.52
     ——
    0.52
     Partly
    0.52
     Kitchen
    0.50
    POSITIVE LOGITS
    newL
    0.66
     recieve
    0.63
     recieved
    0.61
    recieve
    0.60
    ographically
    0.56
     wouldnt
    0.55
    ㅇ
    0.54
     ലഭ
    0.54
    പോലെ
    0.53
    ','"+
    0.52
    Activations Density 0.003%

    No Known Activations