© 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-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 7313
    Prev
    Next
    INDEX
    Explanations

    hell bent or Hellboy

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

    The neuron fires on any occurrence of “hell” (including as part of larger words or titles).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ステム
    -0.80
     topp
    -0.79
    为您
    -0.77
     trám
    -0.77
    ongeza
    -0.76
     話
    -0.75
    礼物
    -0.75
     típicos
    -0.74
     Atlántico
    -0.74
    几点
    -0.72
    POSITIVE LOGITS
    uva
    1.54
    fire
    1.13
     Yeah
    1.10
    raiser
    1.04
     yeah
    1.02
    ishly
    0.99
    kite
    0.98
    Yeah
    0.96
    bent
    0.96
    GMP
    0.96
    Activations Density 0.012%

    No Known Activations