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. 10691
    Prev
    Next
    INDEX
    Explanations

    key phrases related to legal matters or immigration issues

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

    followed by prepositions

    np_acts-logits-general · gemini-2.0-flash

    whilst specific outcomes

    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
     flavorful
    -1.24
     neighborhood
    -1.23
    AVOR
    -1.21
     neighborhoods
    -1.20
    ighborhood
    -1.20
    favorite
    -1.20
    FAVOR
    -1.18
     colorful
    -1.15
    Neighbors
    -1.15
     favorite
    -1.15
    POSITIVE LOGITS
    Whilst
    2.21
     Whilst
    2.13
     whilst
    1.92
    initialise
    1.84
     realise
    1.56
     recognise
    1.53
     specialises
    1.53
     realising
    1.53
     realises
    1.53
     realisation
    1.53
    Activations Density 6.131%

    No Known Activations