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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    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. 5823
    Prev
    Next
    INDEX
    Explanations

    files related to the C++ programming language

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

    cpp code

    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
     Standardi
    -0.57
    tingu
    -0.57
    OrBuilder
    -0.54
    rinfo
    -0.53
    ябре
    -0.52
    xymatrix
    -0.52
    UrlResolution
    -0.51
    xil
    -0.51
    inale
    -0.51
    synthesize
    -0.51
    POSITIVE LOGITS
    cpp
    1.44
     cpp
    1.21
    Cpp
    1.04
     CPP
    1.01
    CPP
    0.97
    Rcpp
    0.65
    cplusplus
    0.62
    hpp
    0.54
    pp
    0.51
     chrétien
    0.50
    Activations Density 0.003%

    No Known Activations