© 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. 10-GEMMASCOPE-RES-131K
    7. 39157
    Prev
    Next
    INDEX
    Explanations

    cyclopri**ine**, hydrochlor**ine**, organochlor**ine**

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

    chemical compounds ending in "ine", particularly alkaloids and pharmaceutical substances.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron flags chemical substance names ending in “-ine.”

    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
    2
    -3.88
    0
    -3.77
    3
    -3.41
    5
    -3.31
    the
    -3.27
    s
    -3.22
    1
    -3.20
    7
    -3.16
    ti
    -3.13
    a
    -2.97
    POSITIVE LOGITS
     univerz
    3.00
     ideje
    2.88
    ญี่ป
    2.78
    ſelves
    2.72
    
    2.69
    ですかね
    2.66
    していますが
    2.64
     continúas
    2.55
     hran
    2.53
     kabát
    2.50
    Activations Density 0.073%

    No Known Activations