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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 3120
    Prev
    Next
    INDEX
    Explanations

    characters that are hybrids or possess dual identities, particularly involving vampires

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

    Punctuation before a new sentence

    np_acts-logits-general · gemini-2.0-flash

    vampire and moonlight

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     <>",
    -0.67
    fjspx
    -0.58
    beforeEach
    -0.57
    ritus
    -0.57
    पया
    -0.56
    jectures
    -0.56
     Jehová
    -0.54
     Jefus
    -0.54
     Económica
    -0.53
     Egli
    -0.52
    POSITIVE LOGITS
     vampire
    1.33
     vampires
    1.24
     Vampire
    1.19
     vamp
    1.19
     Vamp
    1.13
    vampire
    1.12
    Vampire
    1.06
    vamp
    1.05
    Vamp
    1.05
     werewolf
    0.95
    Activations Density 0.178%

    No Known Activations