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. 114519
    Prev
    Next
    INDEX
    Explanations

    punctuation and its context

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

    The highlighted comma is often used at the end of a clause or phrase that provides additional information, explanation, or context, before continuing with the main sentence structure. This punctuation marks a natural break or separation between related but distinct parts of a complex sentence.

    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
     BoxDecoration
    -0.71
     relâche
    -0.70
    MLLoader
    -0.68
    LookAnd
    -0.63
     AssemblyTitle
    -0.63
    AsUp
    -0.61
     AssemblyCompany
    -0.61
    
    -0.59
    findpost
    -0.59
    gól
    -0.58
    POSITIVE LOGITS
     they
    0.30
    micro
    0.30
     lu
    0.29
    γέν
    0.28
     it
    0.28
     micro
    0.27
     kal
    0.27
     vissza
    0.27
     («
    0.26
    Fabricante
    0.26
    Activations Density 0.082%

    No Known Activations