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. 2407
    Prev
    Next
    INDEX
    Explanations

    phrases indicating the discovery or evidence of findings in research studies

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

    found that

    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
    -0.71
     незавершена
    -0.60
    ">//
    -0.60
     betweenstory
    -0.58
     queſta
    -0.57
    🔰
    -0.56
     GK
    -0.55
     OFDb
    -0.54
     Equi
    -0.54
     AssemblyTitle
    -0.54
    POSITIVE LOGITS
     encontraron
    0.63
    พบ
    0.57
     found
    0.56
     ditemukan
    0.54
     encontró
    0.50
    發現
    0.49
     Encuentra
    0.48
     encontrou
    0.48
     encontramos
    0.48
     constater
    0.48
    Activations Density 0.068%

    No Known Activations