© 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. 60640
    Prev
    Next
    INDEX
    Explanations

    verbs followed by prepositions or adjectives

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

    The neuron fires strongly on content verbs—that is, words denoting actions (in their base or infinitive form).

    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
     such
    -1.56
     these
    -1.45
    ebly
    -1.45
     those
    -1.41
     either
    -1.38
    :
    -1.33
     other
    -1.32
     both
    -1.31
     things
    -1.31
     STARTED
    -1.30
    POSITIVE LOGITS
     and
    1.71
     [],
    1.55
    !,
    1.52
    ~,
    1.51
     {},
    1.49
    (),
    1.48
     "",
    1.46
    ?,
    1.44
     '',
    1.39
    ”、“
    1.39
    Activations Density 0.143%

    No Known Activations