© 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-IT
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 8987
    Prev
    Next
    INDEX
    Explanations

    queries related to test case modification and requirements in a software testing context

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

    Letters in questions or options

    np_acts-logits-general · gemini-2.0-flash

    closing parenthesis followed by a word

    np_acts-logits-general · gemini-2.5-flash-lite

    multiple-choice question formatting cues, especially answer choice labels and answer/solution markers.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    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
     BoxFit
    -0.60
    wijl
    -0.50
    zugehen
    -0.47
    tahui
    -0.46
     pokrač
    -0.45
     lisäksi
    -0.44
     gången
    -0.43
     IBOutlet
    -0.43
    RTSN
    -0.42
     потім
    -0.42
    POSITIVE LOGITS
    oredCriteria
    0.44
    RegressionTest
    0.41
     heel
    0.38
    tagHelperRunner
    0.37
    новниш
    0.36
     Walkover
    0.36
    Heel
    0.36
     Priv
    0.35
    CodeAttribute
    0.35
     setAddress
    0.34
    Activations Density 0.223%

    No Known Activations