© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 36230
    Prev
    Next
    INDEX
    Explanations

    end of clauses

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

    The neuron lights up on first‐person descriptions of what the poster has done or set up (e.g. “I …”, “using …”, “implemented …”, “prepared …”)—i.e. narrative of prior steps/troubleshooting.

    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
     if
    -2.03
     might
    -1.63
     will
    -1.58
     more
    -1.53
     or
    -1.49
     would
    -1.43
     may
    -1.42
     если
    -1.37
     maybe
    -1.36
     time
    -1.34
    POSITIVE LOGITS
    successfully
    1.33
     resolução
    1.23
     utilisant
    1.23
     DÍA
    1.23
     但是
    1.22
     黑色
    1.21
    のですが
    1.18
     betreffende
    1.17
    attempt
    1.16
    trying
    1.14
    Activations Density 1.581%

    No Known Activations