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-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 124851
    Prev
    Next
    INDEX
    Explanations

    prepositions and conjunctions indicating comparisons or relations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    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
     heart
    -0.48
     hearts
    -0.38
    Heart
    -0.37
     Inspectors
    -0.36
     East
    -0.35
    heart
    -0.35
    تفصیلات
    -0.35
    ện
    -0.34
     Heart
    -0.34
     Next
    -0.34
    POSITIVE LOGITS
     mentioned
    0.59
    正如
    0.59
     CreateTagHelper
    0.59
     متعلقه
    0.58
    AndEndTag
    0.57
    Ikus
    0.57
    indicated
    0.56
     alluded
    0.54
     disebutkan
    0.53
    agaimana
    0.52
    Activations Density 0.090%

    No Known Activations