© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 21743
    Prev
    Next
    INDEX
    Explanations

    prepositions and auxiliary verbs

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

    The neuron fires strongly on the main result‐ or method‐bearing verbs and their auxiliaries in scientific prose—words like “resulting,” “are,” “provided,” “report,” “failed,” etc., that signal a key finding or procedural step.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    high-frequency function words—especially prepositional and auxiliary/linking terms that connect phrases and mark grammatical relations.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    passive voice constructions, particularly the auxiliary verb "are" in phrases describing states or conditions.

    oai_token-act-pair · claude-4-5-sonnetTriggered 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
     juist
    -1.29
    了过来
    -1.27
     piu
    -1.27
    を変える
    -1.25
     piac
    -1.24
     hanya
    -1.20
    redientes
    -1.19
     basit
    -1.17
     éx
    -1.16
    beruf
    -1.15
    POSITIVE LOGITS
     them
    1.55
     themselves
    1.45
     as
    1.43
     they
    1.39
     various
    1.32
    他们
    1.29
     них
    1.24
    他們
    1.23
     the
    1.21
     mereka
    1.20
    Activations Density 0.338%

    No Known Activations