© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-27B-IT
    3. 31-GEMMASCOPE-2-RES-262K
    4. 57163
    Prev
    Next
    INDEX
    Explanations

    English proper nouns

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

    This neuron detects long runs of alphanumeric/ASCII-encoded data (e.g., base64 or other long encoded/code-like token sequences).

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_31_width_262k_l0_medium
    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
    <unused2114>
    0.07
     които
    0.07
    ཌ
    0.07
     precautions
    0.07
     које
    0.07
    פ
    0.07
    циона
    0.06
         
    0.06
    <tr>
    0.06
    <0xF3>
    0.06
    POSITIVE LOGITS
     Mitochondrial
    0.09
     Orioles
    0.09
    hh
    0.09
     a
    0.09
    ebar
    0.09
    hhhh
    0.09
    utérus
    0.09
    olidated
    0.09
     Legends
    0.09
    hhh
    0.09
    Activations Density 0.000%

    No Known Activations