© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 129710
    Prev
    Next
    INDEX
    Explanations

    The main thing this neuron does is pick out imperative verbs or direct commands (e.g. “go,” “get,” “find,” “skip,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    go [action verb]

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    cún
    -1.08
    敷地
    -1.02
    遑
    -1.02
     flotante
    -0.97
     commuting
    -0.97
    ría
    -0.96
     strany
    -0.93
    Ephesians
    -0.93
     know
    -0.93
     gerekti
    -0.93
    POSITIVE LOGITS
    Už
    1.02
    respectively
    0.96
    italien
    0.92
     menarik
    0.90
     kannst
    0.90
    nother
    0.90
    ⟩
    0.88
    atyw
    0.86
    //};
    0.85
     проє
    0.84
    Activations Density 0.025%

    No Known Activations