© 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. 34-GEMMASCOPE-RES-131K
    7. 48236
    Prev
    Next
    INDEX
    Explanations

    See the License

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

    The neuron fires for numeric tokens—any standalone numbers or digit sequences (e.g. version numbers, IDs, counts).

    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
    行
    -1.11
     –
    -0.91
    ſtra
    -0.82
    來
    -0.74
     perched
    -0.73
     şi
    -0.71
    へ行
    -0.71
    ḻ
    -0.69
    老大
    -0.69
    立
    -0.68
    POSITIVE LOGITS
    「
    1.35
     «
    1.11
    、「
    1.06
    ⼿
    0.97
    ，「
    0.96
     ‘‘
    0.96
    。「
    0.91
     „
    0.83
    ：「
    0.82
     стала
    0.82
    Activations Density 0.247%

    No Known Activations