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-2-9B
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 114635
    Prev
    Next
    INDEX
    Explanations

    references to mathematical operations and expressions

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

    Variables, coefficients, or specialized technical terms that appear in mathematical expressions or scientific contexts, often representing key elements in formulas or domain-specific terminology.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    OGND
    -0.53
    openzeppelin
    -0.50
    httphttps
    -0.47
    Cyfeiriadau
    -0.46
     
    -0.43
    SpringRunner
    -0.43
     CanadaChoose
    -0.42
    IUrlHelper
    -0.42
    הערות
    -0.41
     مق
    -0.40
    POSITIVE LOGITS
    LookAnd
    0.58
    Tikang
    0.47
     wikipagina
    0.42
     ddelweddau
    0.42
    utives
    0.41
    RTLI
    0.41
    الحياه
    0.41
    الدراسه
    0.40
    ⟬
    0.40
    襪
    0.39
    Activations Density 0.752%

    No Known Activations