© 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. 92553
    Prev
    Next
    INDEX
    Explanations

    short words

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

    Explanation of neuron 4 behavior: the main thing this neuron does is find boilerplate license text (e.g. the word “software” in Apache-style headers).

    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
    ='';
    -0.89
    СТВ
    -0.85
    ọi
    -0.84
     plata
    -0.82
     Savannah
    -0.81
     Moraes
    -0.81
     Rusia
    -0.81
    ඝ
    -0.79
     Princesa
    -0.79
     tecn
    -0.77
    POSITIVE LOGITS
     short
    1.30
    Short
    1.23
     Short
    1.16
    short
    0.96
    breviated
    0.91
    SHORT
    0.89
    hort
    0.85
     shorts
    0.82
     rövid
    0.81
     SHORT
    0.80
    Activations Density 0.000%

    No Known Activations