© 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. 10988
    Prev
    Next
    INDEX
    Explanations

    instances of the word "return" and its variants in the context of programming conventions

    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
     AssemblyTitle
    -0.64
    hoeddwyd
    -0.62
     blockSize
    -0.60
     nodeId
    -0.60
    googleapis
    -0.59
    rungsseite
    -0.59
     createdBy
    -0.57
    SharedCtor
    -0.56
    Mast
    -0.56
     courseId
    -0.56
    POSITIVE LOGITS
     return
    1.22
    return
    1.22
     Return
    1.15
    Return
    1.04
     RETURN
    0.99
    RETURN
    0.93
     returns
    0.93
     returned
    0.93
    Returned
    0.85
    returns
    0.84
    Activations Density 0.096%

    No Known Activations