© 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
    2. Gemma-3-27B-IT
    3. 10-GEMMASCOPE-2-TRANSCODER-262K
    4. 47016
    Prev
    Next
    INDEX
    Explanations

    This neuron seems to detect mentions of physical or structural systems and their components, often in scientific or technical contexts. It appears to associate words like 'machinery', 'apparatus', and 'frameworks'.Looking closer at the provided lists:* **MAX_ACTIVATING_TOKENS**: `machinery`, `apparatus`, `machinery`, `inery` (part of machinery), `frameworks`* **TOKENS_AFTER_MAX_ACTIVATING_TOKEN**: `of` (after machinery), `.` (after apparatus), `Performance` (after inery), `computers` (after 'of' which followed 'machinery')* **TOP_ACTIVATING_TEXTS**: contain phrases like "topological machinery", "bureaucratic apparatus", "RNA machinery", "Turbomachinery Performance", "infrastructure", "legal frameworks", "hardware of computers", "physical network".The core theme revolves around mechanisms, systems, and structures, both physical and conceptual.**Explanation:**machinery apparatus frameworks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_10_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
    мам
    0.59
    မ်း
    0.59
    ме
    0.57
    IN
    0.56
    ่า
    0.56
    nění
    0.55
     as
    0.55
     nhạc
    0.55
    ння
    0.53
     "
    0.52
    POSITIVE LOGITS
    et
    0.73
     기반
    0.70
    jekt
    0.64
     invertible
    0.64
    󰡕
    0.63
    Α
    0.61
     demur
    0.60
     घटा
    0.59
     isomorphisms
    0.59
     irland
    0.59
    Activations Density 0.047%

    No Known Activations