© 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. 34-GEMMASCOPE-RES-131K
    7. 91407
    Prev
    Next
    INDEX
    Explanations

    leave alone

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

    The neuron spikes on occurrences of the verb “leave” (and its morphological variants, often in “leave … alone” contexts).

    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
    erover
    -0.84
     nejen
    -0.82
    cioso
    -0.79
    ᗴ
    -0.79
    ligible
    -0.77
     bawat
    -0.76
    瀟
    -0.76
    Morfo
    -0.76
    дали
    -0.75
    Adverti
    -0.75
    POSITIVE LOGITS
     alone
    3.52
    alone
    2.28
     Alone
    2.06
    Alone
    1.92
     ALONE
    1.78
     allein
    1.74
     be
    1.63
     alleine
    1.38
     alene
    1.34
     undisturbed
    1.22
    Activations Density 0.013%

    No Known Activations