© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 102944
    Prev
    Next
    INDEX
    Explanations

    speaking to media outlets

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

    The neuron responds to reporting‐speech contexts—verbs like “confirm,” “tell,” “say,” “interview(ed),” often with “to,” “with,” or a source following (e.g. “confirmed to CNN,” “in an exclusive interview with AFP,” “told reporters”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     aberta
    -1.56
     décès
    -1.51
    ↵↵
    -1.44
    lastung
    -1.40
     ちゃん
    -1.38
     adorno
    -1.36
    wyczaj
    -1.35
    いかがでしたか
    -1.35
     rangka
    -1.34
     vendeur
    -1.33
    POSITIVE LOGITS
     of
    2.33
    々は
    1.53
            
    1.52
    не
    1.45
     เมื่อ
    1.42
    就此
    1.42
    的事情
    1.41
           
    1.41
        
    1.41
    ).
    1.40
    Activations Density 0.031%

    No Known Activations