© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 926
    Prev
    Next
    INDEX
    Explanations

    specific formulations and their optimization in scientific research contexts

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

    scientific research methodologies and technical experimental procedures.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi

    methodological descriptions of experiments or procedures where researchers optimized parameters or conditions to find optimal results.

    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
     purpoſe
    -0.39
    IRUS
    -0.39
     Unicode
    -0.36
     PCP
    -0.35
    PCP
    -0.34
     Pray
    -0.34
     dezelve
    -0.34
    empêcher
    -0.34
    colgante
    -0.34
    Termin
    -0.33
    POSITIVE LOGITS
     experimentation
    0.64
     experimenting
    0.62
     parha
    0.59
     AssemblyCulture
    0.58
     best
    0.57
     experiments
    0.55
    principalColumn
    0.51
     EXPERIMENTS
    0.51
    complexContent
    0.49
    AntiForgeryToken
    0.48
    Activations Density 0.499%

    No Known Activations