© 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. 22-GEMMASCOPE-RES-131K
    7. 67076
    Prev
    Next
    INDEX
    Explanations

    cloud followed by product or service

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

    The neuron consistently activates on every occurrence of the word “Cloud,” flagging mentions of “Cloud” (as in cloud computing/cloud services).

    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
    皷
    -2.69
    ますが
    -2.42
    釹
    -2.33
    ことです
    -2.31
    釤
    -2.27
    1
    -2.23
     zusätzlichen
    -2.22
     yum
    -2.16
    ?’
    -2.16
     nawet
    -2.14
    POSITIVE LOGITS
    ]
    2.89
    ៚
    2.69
    {
    2.52
    雰
    2.50
    鍇
    2.48
    ꠥ
    2.38
    🢒
    2.34
    ...»
    2.30
    
    2.23
    嘤
    2.22
    Activations Density 0.010%

    No Known Activations