© 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. 34-GEMMASCOPE-RES-131K
    7. 75654
    Prev
    Next
    INDEX
    Explanations

    explains something is complicated

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

    The neuron fires strongly on very common, high-frequency function and discourse words (stopwords and filler tokens) rather than content words.

    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
     tonos
    -1.16
    ではなく
    -1.06
     cafetería
    -1.05
     bì
    -1.05
    rouvez
    -1.05
    държа
    -1.05
     kto
    -1.00
     gmina
    -0.99
    xxxxxxxxxxxxxxxx
    -0.99
    -0.98
    POSITIVE LOGITS
     you
    1.87
     your
    1.39
    filepath
    1.28
    Okay
    1.23
    your
    1.20
    You
    1.14
    Assista
    1.13
     непло
    1.09
    めちゃくちゃ
    1.09
     You
    1.07
    Activations Density 0.085%

    No Known Activations