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. 4-GEMMASCOPE-RES-16K
    6. 625
    Prev
    Next
    INDEX
    Explanations

    code elements related to numerical data types and arithmetic operations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.4.hook_resid_post
    Hook Layer
    4
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    mogat
    -0.62
     newBuilder
    -0.60
     szerint
    -0.56
    bardziej
    -0.56
     Josiah
    -0.56
    bili
    -0.55
    a
    -0.55
    ụy
    -0.55
     potrivit
    -0.54
     akad
    -0.54
    POSITIVE LOGITS
     Integer
    1.93
    Integer
    1.89
     integer
    1.82
    integer
    1.68
     integers
    1.60
    getInteger
    1.49
    INTEGER
    1.38
     INTEGER
    1.32
    integers
    1.25
     Inte
    1.15
    Activations Density 0.050%

    No Known Activations