© 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. 74395
    Prev
    Next
    INDEX
    Explanations

    chondrocytes, erythrocytes, osteocytes, oligodendrocytes

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

    chemical and biological terms containing the syllable "ro", particularly in scientific contexts.

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

    The neuron strongly activates on the substring “nitro” (e.g. in chemical names like hexanitro, tetranitro), i.e. it detects nitro-containing compound terms.

    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
    铼
    -2.91
    柺
    -2.89
    騭
    -2.81
    <bos>
    -2.80
     brukt
    -2.72
     they
    -2.67
    茕
    -2.63
     blive
    -2.56
    at
    -2.53
    籥
    -2.50
    POSITIVE LOGITS
    媄
    3.30
     Upon
    3.20
    0
    3.11
     Meanwhile
    3.02
     -
    2.98
    ro
    2.91
     Alongside
    2.91
     Presently
    2.84
     Earlier
    2.84
     Could
    2.84
    Activations Density 0.035%

    No Known Activations