© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 7098
    Prev
    Next
    INDEX
    Explanations

    also or too

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

    The neuron strongly activates on high-level discourse tokens—especially sentence-initial and emphatic connectors and pronouns such as “So,” “so,” “I,” “We,” etc.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    𝑏
    0.60
     همچنین
    0.59
    வதை
    0.57
    फु
    0.52
    ………………………………
    0.52
    🟣
    0.52
    했으며
    0.51
    ப்பதை
    0.50
    ുവെ
    0.50
     மேலும்
    0.49
    POSITIVE LOGITS
     too
    4.16
    too
    3.64
     TOO
    3.42
     Too
    3.28
    Too
    3.27
     тоже
    3.17
     слишком
    2.75
    太
    2.75
     demasiado
    2.67
    TOO
    2.63
    Activations Density 0.268%

    No Known Activations