© 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. 34-GEMMASCOPE-RES-131K
    4. 98407
    Prev
    Next
    INDEX
    Explanations

    absorption and adsorption

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

    words related to absorption or adsorption properties of materials.

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

    The neuron activates on occurrences of the roots “absorb” or “adsorb” (e.g. absorbent, absorptive, adsorption), i.e. words describing absorption/adsorption processes or materials.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     рисунок
    -0.88
     isn
    -0.85
    mei
    -0.84
     aprobación
    -0.84
    锢
    -0.83
     хвилин
    -0.82
     واش
    -0.81
     秒
    -0.81
     واضح
    -0.81
    рения
    -0.80
    POSITIVE LOGITS
    tion
    0.98
    anía
    0.84
     banish
    0.82
     eliminate
    0.81
    Axiom
    0.80
     Hannibal
    0.77
    究極
    0.77
     shocks
    0.75
     prevent
    0.75
    周泽
    0.75
    Activations Density 0.034%

    No Known Activations