Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 50451
    Prev
    Next
    INDEX
    Explanations

    references to angels and angelic imagery

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    angels and hell

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     nahilalakip
    -0.51
    Havolalar
    -0.49
     Lordship
    -0.49
    modelBuilder
    -0.49
    Owner
    -0.49
     saveiro
    -0.48
     nových
    -0.47
     lordship
    -0.47
     nového
    -0.46
     pinulongan
    -0.46
    POSITIVE LOGITS
     angels
    0.81
     Angels
    0.74
     angel
    0.72
    Angels
    0.68
    👼
    0.66
     анге
    0.64
    angels
    0.64
     angelic
    0.62
     Angel
    0.60
    天使
    0.54
    Activations Density 0.015%

    No Known Activations