© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 129201
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on discourse-level concession or context markers—words like “considering,” “given (that),” “since,” etc., that introduce a subordinate clause providing background or caveat.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    considering that

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     good
    -1.70
     will
    -1.55
     strange
    -1.52
     give
    -1.49
     different
    -1.48
     various
    -1.47
     with
    -1.45
     allow
    -1.45
     through
    -1.40
     helps
    -1.38
    POSITIVE LOGITS
    basically
    1.68
     praticamente
    1.53
     stiller
    1.52
    almost
    1.52
     forment
    1.52
     nogal
    1.51
     geene
    1.49
     haast
    1.48
     insanely
    1.48
    
    1.48
    Activations Density 0.020%

    No Known Activations