Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 54358
    Prev
    Next
    INDEX
    Explanations

    declarations and assignments of string variables in coding contexts

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

    string declarations and usage

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     يتيمه
    -0.69
     disambiguazione
    -0.57
    tagHelperRunner
    -0.51
     EconPapers
    -0.49
    enumii
    -0.49
    httphttps
    -0.49
     Normdatei
    -0.45
     pinulongan
    -0.45
     AssemblyCompany
    -0.45
     kasarigan
    -0.45
    POSITIVE LOGITS
     createSprite
    0.44
    addImage
    0.42
    Rüyada
    0.41
    nologue
    0.41
    انتهای
    0.40
    GOTREF
    0.38
     autorytatywna
    0.38
    лтемелер
    0.37
    Interviews
    0.35
    GeneratedMessage
    0.35
    Activations Density 0.007%

    No Known Activations