© 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. 8025
    Prev
    Next
    INDEX
    Explanations

    references to parameters in contexts related to algorithms or computational methods

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
    WithMany
    -0.50
    StructEnd
    -0.48
    ../../
    -0.48
    wakili
    -0.41
    <bos>
    -0.41
    Wel
    -0.39
    guilty
    -0.38
    ../../../
    -0.38
    OrWhiteSpace
    -0.38
    ="#"><
    -0.37
    POSITIVE LOGITS
     Param
    1.16
     Parameter
    1.14
     parameter
    1.09
     Parameters
    1.09
    Parameter
    1.05
     parameters
    1.05
     param
    1.02
    Parameters
    1.01
    PARAM
    0.97
     PARAM
    0.97
    Activations Density 0.086%

    No Known Activations