© 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
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 6999
    Prev
    Next
    INDEX
    Explanations

    The neuron selectively activates on imperative cooking steps—i.e. recipe instruction verbs like “heat,” “add,” “mix,” etc.

    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
     bowl
    0.83
     bowls
    0.79
    bowl
    0.77
     cake
    0.74
    묽
    0.73
    cold
    0.72
     porridge
    0.72
     boiled
    0.71
    boiled
    0.71
     curd
    0.70
    POSITIVE LOGITS
     spice
    0.60
     Spice
    0.59
     zemlje
    0.55
     Ев
    0.54
    ับสนุน
    0.54
     judíos
    0.54
     евре
    0.53
     eleições
    0.53
    ])).
    0.52
     Bevölker
    0.52
    Activations Density 0.499%

    No Known Activations