© 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. 22-GEMMASCOPE-RES-131K
    4. 13872
    Prev
    Next
    INDEX
    Explanations

    a followed by a description

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

    This neuron detects the idiomatic construction “It’s a …,” especially “It’s a” followed by an adjective (e.g. “It’s a great,” “It’s a common,” “It’s a tricky,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     this
    -1.63
     its
    -1.59
     in
    -1.59
    																				
    -1.13
     as
    -1.12
     it
    -1.12
     with
    -1.09
    													
    -1.09
    																					
    -1.06
    																		
    -1.03
    POSITIVE LOGITS
     الذين
    1.16
     moment
    1.08
     time
    1.03
    They
    1.02
     they
    1.02
     tactic
    1.00
     matter
    0.98
    llium
    0.98
     amplia
    0.98
     TextInputType
    0.98
    Activations Density 0.033%

    No Known Activations