© 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
    2. Gemma-3-27B-IT
    3. 37-GEMMASCOPE-2-TRANSCODER-262K
    4. 203088
    Prev
    Next
    INDEX
    Explanations

    physics, computer programming, earning money, PhD, quantum gravity, General Relativity, theoretical physics, CERN, effective field theory, Generalized Uncertainty Principle, holographic principle, Feynman rules, quantum field theory, topology, GR, geometry, mathematical underpinnings, realism, affiliation, Institute for Advanced Study, Princeton, string theory.**Analysis:*** **MAX_ACTIVATING_TOKENS** and **TOKENS_AFTER_MAX_ACTIVATING_TOKEN**: These show tokens related to academic/technical contexts (`affiliation`, `Stanford`, `quantum`, `0`) and some punctuation. The sequence `quantum` appearing after `the` or `let` is interesting.* **TOP_POSITIVE_LOGITS**: These are diverse but include scientific journal names (`JHEP`) and terms that *sound* technical or potentially related to systems/fields (`Beam`, `beam`, `Classical`). `الجما` and `جما` are Arabic for "community" or "group," which could relate to academic communities. `ሰ` is an Ethiopic character. `antis` might relate to anti-matter or related physics concepts.* **TOP_ACTIVATING_TEXTS**: This is the most informative. * "fastest way to earn enough money to stop working" (with a PhD in supergravity/programming) - this seems to be a user query, not the neuron's *detection*.**quantum physics and affiliations**

    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_37_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
     Lipstick
    0.45
     projections
    0.44
    atrice
    0.40
     matematika
    0.40
    orkan
    0.39
     мото
    0.38
     mayhem
    0.38
    範囲
    0.38
    libc
    0.37
    ത്വം
    0.37
    POSITIVE LOGITS
     dole
    0.42
     الجما
    0.42
     endowed
    0.41
     JHEP
    0.39
     جما
    0.38
    Beam
    0.37
     beam
    0.37
     ሰ
    0.36
     antis
    0.36
    Classical
    0.36
    Activations Density 0.000%

    No Known Activations