© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 571
    Prev
    Next
    INDEX
    Explanations

    creating and reacting chemical groups

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

    chemical reaction and molecular structure descriptions.

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

    The neuron fires on phrases describing the physical folding or wrapping of layered materials (e.g. sheets folding along an axis or layers wrapping onto themselves).

    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
    allclose
    -0.98
     álbum
    -0.96
     almohada
    -0.91
    ッション
    -0.88
    ushort
    -0.86
     portfolio
    -0.85
    ンク
    -0.85
     infants
    -0.83
     estarán
    -0.82
     mieć
    -0.82
    POSITIVE LOGITS
     δημιουργ
    1.15
     creating
    0.94
     producing
    0.91
     suffisamment
    0.91
     construir
    0.90
    สร้าง
    0.89
     DIFFERENT
    0.86
     creando
    0.85
    creating
    0.85
    mith
    0.85
    Activations Density 0.173%

    No Known Activations