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. 9144
    Prev
    Next
    INDEX
    Explanations

    legal terminology related to doctrines and court rulings

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
    ⓧ
    -0.77
     avoient
    -0.77
     تضيفلها
    -0.74
     كومونز
    -0.73
     enfans
    -0.72
     ModelExpression
    -0.71
     zijne
    -0.70
    Personensuche
    -0.69
    OGND
    -0.69
     dieux
    -0.69
    POSITIVE LOGITS
     concept
    1.57
     approach
    1.36
     idea
    1.34
     phenomenon
    1.20
     scenario
    1.19
     theory
    1.19
     strategy
    1.13
     notion
    1.12
     doctrine
    1.08
     syndrome
    1.07
    Activations Density 1.186%

    No Known Activations