© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 31-GEMMASCOPE-RES-16K
    11. 2647
    Prev
    Next
    INDEX
    Explanations

    programming language syntax related to function definitions and expected outcomes from function calls

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

    arrow function syntax and test assertion patterns in code.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    didSet
    -0.47
    npmjs
    -0.45
    MLLoader
    -0.45
    Халык
    -0.42
     بست
    -0.41
     utafitiHapana
    -0.41
     jButton
    -0.41
     RemoteException
    -0.40
     तैयार
    -0.40
     BorderLayout
    -0.40
    POSITIVE LOGITS
     expect
    1.22
     Expect
    1.17
     assert
    1.10
     expecting
    1.09
     expectation
    1.09
    expect
    1.08
     expects
    1.05
     expected
    1.04
    Expect
    1.04
    Assert
    1.00
    Activations Density 0.227%

    No Known Activations