© 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. 7222
    Prev
    Next
    INDEX
    Explanations

    terms and symbols related to mathematical or scientific constructs

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

    programming code syntax with bitfield notation (specifically detecting colon-semicolon patterns in source code).

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered 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
    SequentialGroup
    -0.55
     kasarigan
    -0.55
    jsxFileName
    -0.54
    fromnode
    -0.53
    didSet
    -0.50
    __(/*!
    -0.49
    Hentet
    -0.49
     <=",
    -0.48
    complexContent
    -0.48
    RenderAtEndOf
    -0.48
    POSITIVE LOGITS
     myſelf
    0.47
     itſelf
    0.45
    økt
    0.45
     ſeveral
    0.42
     paisajes
    0.41
    ésor
    0.41
    veyard
    0.40
     themſelves
    0.40
     Drexel
    0.40
    César
    0.40
    Activations Density 0.867%

    No Known Activations