© 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. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 10433
    Prev
    Next
    INDEX
    Explanations

    references to character relationships and dynamics

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

    "Who" followed by a verb

    np_acts-logits-general · gemini-2.0-flash

    people who like or know things

    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
    pons
    -0.54
     zostało
    -0.53
    ksikon
    -0.51
    Viitteet
    -0.51
     getRule
    -0.51
    ณ
    -0.50
     zostały
    -0.47
    Secure
    -0.46
    EndContext
    -0.46
    stantial
    -0.45
    POSITIVE LOGITS
     prefers
    1.39
     loves
    1.33
     prefer
    1.28
     likes
    1.27
     preferring
    1.22
    prefer
    1.19
     hates
    1.19
    Prefer
    1.17
     Prefer
    1.15
     gusta
    1.08
    Activations Density 0.355%

    No Known Activations