© 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-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 6000
    Prev
    Next
    INDEX
    Explanations

    code and references specific to programming languages, particularly Python and related technologies

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

    the programming language "Python."

    np_acts-logits-general · gemini-2.0-flash

    Python programming language

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 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
    TagHelper
    -0.72
    ruptcy
    -0.62
    ViewImports
    -0.60
    }\]
    -0.52
    AnchorStyles
    -0.52
    Initializable
    -0.51
    writerow
    -0.50
    azgo
    -0.49
    település
    -0.48
    ViewType
    -0.48
    POSITIVE LOGITS
     version
    0.69
    based
    0.65
     native
    0.63
     based
    0.60
     linkovi
    0.60
     versions
    0.60
    native
    0.58
     gyhoeddwyd
    0.57
     programming
    0.56
     "..\..\
    0.56
    Activations Density 0.600%

    No Known Activations