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-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 64749
    Prev
    Next
    INDEX
    Explanations

    verbs and adjectives indicating processes or actions in various contexts

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

    creation, testing, and discovery

    np_acts-logits-general · gemini-2.5-flash-lite
    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
      
    -0.40
     during
    -0.38
    ↵↵
    -0.37
    a
    -0.35
    <eos>
    -0.32
     was
    -0.31
     durante
    -0.31
     could
    -0.31
    for
    -0.30
     the
    -0.30
    POSITIVE LOGITS
     nakalista
    1.03
     propOrder
    1.02
    <unused79>
    1.01
    [@BOS@]
    1.00
    <unused42>
    1.00
    <unused16>
    1.00
    <unused41>
    1.00
    <unused28>
    1.00
    <unused8>
    1.00
    <unused14>
    1.00
    Activations Density 0.155%

    No Known Activations