© 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. 55570
    Prev
    Next
    INDEX
    Explanations

    introduces definitions or conditions

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

    The neuron activates primarily on numeric literals (digit strings) in the text—that is, it detects numbers.

    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
    que
    -0.88
     nor
    -0.87
     $-$
    -0.86
    ᴱ
    -0.84
    fi
    -0.84
    atique
    -0.84
     potrebbero
    -0.81
    q
    -0.81
    \}=\
    -0.81
    nullable
    -0.81
    POSITIVE LOGITS
     where
    4.53
    where
    2.66
    Where
    2.63
     где
    2.58
     où
    2.48
     Where
    2.47
     with
    2.36
     gdzie
    2.30
     donde
    2.22
     όπου
    2.22
    Activations Density 0.081%

    No Known Activations