© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 72322
    Prev
    Next
    INDEX
    Explanations

    words starting with bu

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

    The neuron is highly activated by the character sequence “bu,” i.e. it detects tokens beginning with “bu.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     hochwertige
    -2.58
    荭
    -2.48
    idk
    -2.44
     réalisés
    -2.44
    ere
    -2.39
     thee
    -2.33
    夸张
    -2.33
    咵
    -2.27
    
    -2.27
    als
    -2.22
    POSITIVE LOGITS
    鹚
    3.05
    鲼
    3.02
    逦
    2.91
    ClassNotFound
    2.41
    怍
    2.36
    蚣
    2.28
    ፫
    2.28
    "]=
    2.28
    thenburg
    2.27
    )?
    2.25
    Activations Density 0.018%

    No Known Activations