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. 14285
    Prev
    Next
    INDEX
    Explanations

    references to television news networks and platforms

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

    Heisenberg and news outlets

    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
     يتيمه
    -0.67
     AssemblyCulture
    -0.65
     nakalista
    -0.56
    Clik
    -0.55
    LookAnd
    -0.54
     Infórmanos
    -0.54
    AutoresizingMask
    -0.54
    providedIn
    -0.54
    rshire
    -0.52
    oader
    -0.52
    POSITIVE LOGITS
     ESPN
    1.31
    ESPN
    1.20
    espn
    1.07
     Disney
    0.65
    Disney
    0.61
     esp
    0.57
     ESP
    0.56
     Deportes
    0.54
    EventName
    0.53
     CNN
    0.52
    Activations Density 0.003%

    No Known Activations