© 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. 34-GEMMASCOPE-RES-131K
    7. 32734
    Prev
    Next
    INDEX
    Explanations

    adding information to speech

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

    The neuron fires on the comma that introduces a reporter‐attribution phrase (e.g. “, says,” “, adding,” “, who,” 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_34/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
     said
    -1.19
     that
    -1.16
     this
    -1.10
     SAYS
    -1.08
    said
    -1.05
     these
    -1.05
    Says
    -1.03
    says
    -1.01
     Says
    -0.99
     situaciones
    -0.99
    POSITIVE LOGITS
     adding
    1.63
     referring
    1.52
     speaking
    1.34
     noting
    1.24
    refer
    1.18
    Доба
    1.17
     “
    1.16
     citing
    1.11
     ajout
    1.08
     opět
    1.08
    Activations Density 0.011%

    No Known Activations