© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 37945
    Prev
    Next
    INDEX
    Explanations

    states of being

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

    The neuron strongly activates on second‐person address words and related auxiliaries—tokens like “you,” “are,” “is,” “not,” etc., that directly address the reader.

    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
     theſe
    -1.18
    更大的
    -1.12
    nappropriate
    -1.08
    衕
    -1.05
     robustness
    -1.04
    superior
    -1.02
     poorer
    -1.02
    ícias
    -1.01
     takiego
    -1.01
     robusto
    -1.00
    POSITIVE LOGITS
    more
    1.00
     Extremely
    0.98
     más
    0.98
     Moderately
    0.95
     chagrin
    0.90
    
    0.90
    มีความ
    0.88
    both
    0.88
    足够
    0.87
     extremely
    0.86
    Activations Density 0.163%

    No Known Activations