© 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 - 131k
    9. 31-GEMMASCOPE-RES-131K
    11. 6461
    Prev
    Next
    INDEX
    Explanations

    instances of formatting commands and document structure in a programming or markup language context

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

    latex: mkern commands

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     querían
    -0.33
     próprias
    -0.30
    CodeAttribute
    -0.29
     arque
    -0.28
    さえ
    -0.27
    openSession
    -0.27
     चाहते
    -0.27
     próprios
    -0.26
     ventilación
    -0.26
    nesc
    -0.26
    POSITIVE LOGITS
    CloseOperation
    0.60
    principalTable
    0.59
     papers
    0.58
    InputTagHelper
    0.57
    出版年
    0.56
    yntaxException
    0.56
     papiers
    0.56
     nahilalakip
    0.54
    enterOuterAlt
    0.54
     getTitle
    0.53
    Activations Density 0.004%

    No Known Activations