© 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. 32271
    Prev
    Next
    INDEX
    Explanations

    burst into, broke down

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

    The neuron detects words and phrases describing someone breaking down in tears or becoming emotionally overwhelmed.

    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
     vettore
    -0.84
     llega
    -0.80
     także
    -0.76
    龇
    -0.75
     bahkan
    -0.75
    γεν
    -0.74
     mengin
    -0.74
     parfois
    -0.74
    nib
    -0.73
    幸いです
    -0.73
    POSITIVE LOGITS
     burst
    3.38
    burst
    2.56
    Burst
    2.48
     Burst
    2.47
     bursting
    2.33
     bursts
    2.27
     start
    1.93
     started
    1.85
     begin
    1.84
    bursts
    1.83
    Activations Density 0.025%

    No Known Activations