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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 16376
    Prev
    Next
    INDEX
    Explanations

    specific character traits and actions associated with main characters in a narrative

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

    Possessive nouns describing people in servitude

    np_acts-logits-general · gemini-2.0-flash

    possessives followed by associates

    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
    BeginContext
    -0.64
    DockStyle
    -0.61
    PerformLayout
    -0.56
     Grandfather
    -0.54
     grandfather
    -0.54
     reden
    -0.51
     quán
    -0.51
     Parent
    -0.50
    BuildContext
    -0.50
     parent
    -0.48
    POSITIVE LOGITS
     subordinates
    1.29
     assistants
    1.26
     assistant
    1.16
     subordinate
    1.11
     followers
    1.09
     disciples
    1.07
     minions
    1.04
     disciple
    1.01
     Assistants
    0.99
     loyal
    0.99
    Activations Density 0.513%

    No Known Activations