© 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. 53878
    Prev
    Next
    INDEX
    Explanations

    maximize and minimize

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

    This neuron tracks how far into the segment a token appears, with its activation rising steadily for words later in the text.

    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
    곤
    -0.84
     easier
    -0.82
    じて
    -0.81
    abilidade
    -0.79
    StatusLabel
    -0.77
     Laine
    -0.75
    setViewName
    -0.75
     utilising
    -0.75
    更容易
    -0.74
    Exited
    -0.74
    POSITIVE LOGITS
     maximize
    3.86
    maxim
    3.72
    Maxim
    3.50
     maximizing
    3.44
     maxim
    3.41
     maximization
    3.39
     maximise
    3.38
     Maxim
    3.34
     maximizes
    3.27
    maximize
    3.16
    Activations Density 0.117%

    No Known Activations