© 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. 84385
    Prev
    Next
    INDEX
    Explanations

    introduces additional information

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

    The neuron detects the discourse adverb “also” (and its immediate superlative/adverbial constructions, e.g. “also considered,” “also the highest in”) marking additive or emphatic statements.

    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
     especially
    -1.27
     camadas
    -1.18
     even
    -1.13
     instead
    -1.07
     mainly
    -1.06
     ApiException
    -1.05
     такі
    -1.05
    Specifically
    -1.04
     however
    -1.04
     such
    -1.04
    POSITIVE LOGITS
     also
    1.16
    F
    1.09
     também
    1.09
    b
    1.02
    g
    0.97
    G
    0.96
     serves
    0.94
    m
    0.93
     غذا
    0.91
     uten
    0.91
    Activations Density 0.041%

    No Known Activations