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. 92206
    Prev
    Next
    INDEX
    Explanations

    keywords related to auditing, oversight, and regulatory processes

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

    TV commercials, projection, pressing, delivery, bond, groups, method

    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
    rungsseite
    -0.75
     doInBackground
    -0.58
     nakalista
    -0.53
    TestTools
    -0.51
    libft
    -0.49
     XCTest
    -0.48
    ContextHolder
    -0.47
     intStringLen
    -0.47
    ########.
    -0.47
     snippetHide
    -0.46
    POSITIVE LOGITS
    thâu
    0.47
     ModelExpression
    0.42
     jednocześnie
    0.42
     doporu
    0.42
     samtidig
    0.41
    eseorang
    0.40
    empereur
    0.40
     allAfrica
    0.40
    vorschaubild
    0.40
     stället
    0.40
    Activations Density 0.167%

    No Known Activations