© 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-3-27B-IT
    5. 40-GEMMASCOPE-2-TRANSCODER-262K
    7. 0
    Prev
    Next
    INDEX
    Explanations

    units of measurement (mg, ml, IE, Eq)

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

    units of measurement **Method used:** 3. The TOP_POSITIVE_LOGITS all relate to "IU" (International Units) and "Units," indicating the neuron predicts measurement units, especially in medical/scientific contexts shown in the activating texts

    np_max-act-logits · claude-4-5-haikuTriggered by @enricobottazzi

    numbers or units of measurement in medical or scientific contexts.

    oai_token-act-pair · gemini-2.5-flashTriggered by @enricobottazzi
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_40_width_262k_l0_small_affine
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     híbr
    0.41
    eren
    0.40
     Hybrid
    0.39
     hybrid
    0.38
     !=
    0.37
    ुलकर
    0.36
     নির্ধার
    0.36
     exercice
    0.36
     hydraulics
    0.36
    hybrid
    0.35
    POSITIVE LOGITS
     IU
    0.70
    IU
    0.64
     Unit
    0.58
    Units
    0.56
     International
    0.55
     Units
    0.55
    Unit
    0.52
     单位
    0.52
    单位
    0.52
     यूनिट
    0.52
    Activations Density 0.046%

    No Known Activations