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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 18-GEMMASCOPE-RES-16K
    6. 692
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions and formulas related to statistical concepts

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

    Expressions with square roots

    np_acts-logits-general · gemini-2.0-flash

    mathematical square roots

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     hos
    -0.62
    ImageContext
    -0.54
     Empty
    -0.52
    パート
    -0.49
    XMLSchema
    -0.47
     empty
    -0.46
    expandindo
    -0.44
     host
    -0.43
    Fone
    -0.43
    ThroughAttribute
    -0.43
    POSITIVE LOGITS
    sqrt
    1.59
     sqrt
    1.50
     √
    1.30
    Sqrt
    1.27
    √
    1.26
     Root
    1.03
     root
    1.03
    qrt
    0.99
    Root
    0.97
    root
    0.93
    Activations Density 1.010%

    No Known Activations