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. 73756
    Prev
    Next
    INDEX
    Explanations

    complex technical terms and mathematical variables related to data analysis and modeling

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

    Important elements in scientific and technical texts, including variable references, mathematical symbols, punctuation in equations, and 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
    MENAFN
    -0.43
     nakalista
    -0.43
    pædia
    -0.40
    
    -0.39
    shtml
    -0.38
    Pref
    -0.38
     ویکی‌پدیای
    -0.38
    ኛ
    -0.36
     pinulongan
    -0.36
     iconLine
    -0.36
    POSITIVE LOGITS
    WriteTagHelper
    0.49
    RTGC
    0.49
    ByVersion
    0.48
    Aholisi
    0.48
    ніципалі
    0.46
     ब्रेकडाउन
    0.46
    expandindo
    0.43
     eventdata
    0.43
     fotografico
    0.42
    izr
    0.42
    Activations Density 0.087%

    No Known Activations