© 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. 5129
    Prev
    Next
    INDEX
    Explanations

    elements and patterns related to buffer management and memory allocation in programming

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

    low-level programming constructs like buffer lengths, memory allocation, and NULL termination.

    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
    Fav
    -0.40
     help
    -0.38
    argmin
    -0.38
     Jordan
    -0.37
     Clarendon
    -0.37
    ولوجيا
    -0.37
    Jordan
    -0.37
    TypedDataSet
    -0.36
    cshtml
    -0.36
    bin
    -0.36
    POSITIVE LOGITS
    RenderAtEndOf
    0.68
     AssemblyCulture
    0.65
    sizeCache
    0.63
    mybatisplus
    0.63
    unknownFields
    0.59
    OGND
    0.57
    Хьажоргаш
    0.52
    хьтан
    0.50
     ProtoMessage
    0.50
    InputBorder
    0.50
    Activations Density 0.160%

    No Known Activations