© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 66012
    Prev
    Next
    INDEX
    Explanations

    gas, solid, orally, i.v

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

    chemical and physical state notations (like gas, solid, liquid) and scientific formula elements.

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

    The neuron fires on mentions of physical states or phase labels in chemical contexts (e.g. “(g)”, “gases”, “solid(s)”).

    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
     of
    -2.00
    にあります
    -1.33
     filós
    -1.21
    ull
    -1.19
    をします
    -1.18
     that
    -1.17
    さぁ
    -1.15
    来週
    -1.15
    BindingResult
    -1.14
    ℴ
    -1.14
    POSITIVE LOGITS
    そうですが
    1.24
    一品
    1.22
    資料來源
    1.13
    <bos>
    1.12
    тів
    1.10
     насељу
    1.09
    توضیحات
    1.08
     作り方
    1.08
     وتح
    1.08
    அ
    1.05
    Activations Density 0.007%

    No Known Activations