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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 6175
    Prev
    Next
    INDEX
    Explanations

    references to paper-based methods and documentation

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

    keywords related to paper documents and manual processes.

    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
    IntoConstraints
    -0.40
    évaluateur
    -0.39
     vacances
    -0.35
    Boas
    -0.35
     vanguardia
    -0.34
     espec
    -0.32
     Spruce
    -0.32
    InputTagHelper
    -0.32
    حياته
    -0.31
     spes
    -0.31
    POSITIVE LOGITS
     manually
    0.91
     manual
    0.87
    手动
    0.83
     manualmente
    0.81
    Manual
    0.80
    manual
    0.78
     manuelle
    0.77
     Manual
    0.74
    MANUAL
    0.73
     trady
    0.72
    Activations Density 0.527%

    No Known Activations