© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 4525
    Prev
    Next
    INDEX
    Explanations

    technical terms and class structures related to programming documentation

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

    HTML or XML markup tags and code structure elements.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     dass
    -0.28
    RequestId
    -0.26
     anim
    -0.25
    leih
    -0.25
     rewards
    -0.25
     output
    -0.25
    getInfo
    -0.25
     pogod
    -0.25
     being
    -0.24
     ello
    -0.24
    POSITIVE LOGITS
    ScopeManager
    8.25
    SourceChecksum
    1.79
     CreateTagHelper
    1.37
    TagHelper
    1.24
     wireType
    1.22
    fjspx
    1.22
    SequentialGroup
    1.17
    RuleContext
    1.13
    webElementXpaths
    1.09
     estekak
    1.06
    Activations Density 0.001%

    No Known Activations