© 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. 10-GEMMASCOPE-RES-131K
    4. 11535
    Prev
    Next
    INDEX
    Explanations

    changing direction or scale

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

    The neuron fires on dynamic verbs that describe movement or change along a scale (e.g. get, shrink, go, progress, decrease).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     on
    -1.62
     all
    -1.55
     or
    -1.55
    :
    -1.52
     did
    -1.51
     by
    -1.43
     no
    -1.42
     every
    -1.41
     Anderson
    -1.40
     be
    -1.38
    POSITIVE LOGITS
    ፱
    1.38
    icto
    1.38
     incluindo
    1.38
    
    1.38
    etui
    1.36
     than
    1.36
     عوض
    1.35
    やっと
    1.34
    including
    1.34
     blatant
    1.34
    Activations Density 0.040%

    No Known Activations