© 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. 10-GEMMASCOPE-RES-131K
    7. 40352
    Prev
    Next
    INDEX
    Explanations

    ferro words, names, uneasy calm, Lorenzo

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

    The neuron fires on the scientific prefix “ferro” (e.g. in ferro-electric, ferro-magnetic, ferrofluidic, etc.).

    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
    全力
    -2.27
     underscores
    -2.19
    了自己的
    -2.16
    了他的
    -2.13
     
    -2.11
     тех
    -2.09
     “
    -2.09
     Только
    -2.06
     sickly
    -2.06
     Instead
    -2.05
    POSITIVE LOGITS
     />
    2.70
    饪
    2.61
    圌
    2.59
    
    2.59
    但
    2.48
    但我
    2.44
    ൎ
    2.41
    ]
    2.36
     REPRESENTATIVES
    2.33
    carnated
    2.30
    Activations Density 0.005%

    No Known Activations