© 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. 24-GEMMASCOPE-2-RES-16K
    7. 15161
    Prev
    Next
    INDEX
    Explanations

    technical computing or finance terms

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

    The neuron fires on formal, often third-person verbs (e.g. “does,” “displays,” “guarantees,” “prepared”) typical of corporate or legal-style prose.

    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_24_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
     Dee
    0.89
     Бі
    0.80
    u
    0.78
    Dee
    0.75
     Ско
    0.73
    ен
    0.73
    Shock
    0.72
    vanje
    0.71
     Ayrıca
    0.70
    ž
    0.70
    POSITIVE LOGITS
     brevi
    0.81
    мили
    0.78
     métodos
    0.75
    を残
    0.75
     ढंग
    0.74
     calculi
    0.74
     miRNAs
    0.74
     distintos
    0.73
     bushels
    0.72
     chromatography
    0.71
    Activations Density 0.000%

    No Known Activations