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. 8922
    Prev
    Next
    INDEX
    Explanations

    instances of non-verbal communication, particularly when addressing the user or providing documentation

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

    academic references and footnotes

    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
    GTCX
    -0.70
    Rüyada
    -0.49
    JspWriter
    -0.47
    NavController
    -0.45
    OGND
    -0.45
     للاسماء
    -0.45
    TestingModule
    -0.44
    stdc
    -0.44
    OCCURRED
    -0.43
     <=",
    -0.42
    POSITIVE LOGITS
    AnchorTagHelper
    0.47
    Notes
    0.46
     кру
    0.43
     рас
    0.43
    Prompt
    0.42
     Notes
    0.41
    Пор
    0.41
     Reminders
    0.41
    Alert
    0.40
    Staff
    0.40
    Activations Density 0.031%

    No Known Activations