© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 97952
    Prev
    Next
    INDEX
    Explanations

    tree and its contexts

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

    The neuron fires on mentions of “tree” (and its plural or compound forms), i.e. references to trees or tree‐related concepts.

    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
    嵜
    -0.91
    rewsbury
    -0.89
    Griff
    -0.80
     queens
    -0.80
     skit
    -0.79
    Jobs
    -0.78
    richlet
    -0.78
    ↵
    -0.78
    Unis
    -0.77
    背中
    -0.76
    POSITIVE LOGITS
    hug
    1.16
     trunks
    1.05
     hug
    1.03
    🌳
    1.02
    house
    1.02
     species
    1.01
    tops
    0.98
    mendous
    0.98
     stump
    0.92
     lær
    0.92
    Activations Density 0.009%

    No Known Activations