© 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. 36699
    Prev
    Next
    INDEX
    Explanations

    code comments and includes

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

    The neuron flags comment‐style delimiters and preprocessor or import directives (e.g. /** … */, //, #region/#endregion, #include, #ifdef).

    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
     after
    -1.61
     during
    -1.59
     when
    -1.59
     if
    -1.53
     any
    -1.52
     one
    -1.51
     while
    -1.46
     around
    -1.31
     by
    -1.30
     only
    -1.28
    POSITIVE LOGITS
     flera
    1.25
    usually
    1.20
    sorry
    1.20
     FIVE
    1.11
     normalmente
    1.09
     frequentemente
    1.08
     nästan
    1.08
     principalmente
    1.07
     fyra
    1.07
    required
    1.05
    Activations Density 0.036%

    No Known Activations