© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 66209
    Prev
    Next
    INDEX
    Explanations

    and followed by a verb

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

    The neuron consistently flags wording that marks a key function, state, or turning‐point in a description—words like “main,” “focus,” “visible,” “closed,” “locked,” and transition markers such as “but,” “while,” and “also.”

    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
     princípios
    -0.96
    bdb
    -0.94
    seamnă
    -0.91
    場所で
    -0.91
    場所に
    -0.89
     makkelijk
    -0.89
     Вин
    -0.89
     começa
    -0.88
     perdeu
    -0.88
    垭
    -0.88
    POSITIVE LOGITS
    0.92
    ▇
    0.88
    只
    0.88
                
    0.87
    0.86
     surfaces
    0.85
    ✫
    0.85
     differentially
    0.84
     ACTUALLY
    0.84
     asses
    0.84
    Activations Density 0.066%

    No Known Activations