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 - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 15631
    Prev
    Next
    INDEX
    Explanations

    C/C++ preprocessor directives and include statements in code

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

    C++ header files being included

    np_acts-logits-general · gemini-2.0-flash

    programming includes

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76
    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
     letz
    -0.48
     BitmapFactory
    -0.42
     aikana
    -0.42
     løpet
    -0.38
     jähr
    -0.38
     šķ
    -0.37
     laikā
    -0.35
    樣子
    -0.35
     laatste
    -0.35
     그녀
    -0.34
    POSITIVE LOGITS
     initComponents
    0.77
    expandindo
    0.65
     propOrder
    0.65
     autorytatywna
    0.61
     kasarigan
    0.60
    verifyException
    0.57
    aute
    0.55
     headers
    0.55
     Administrativna
    0.55
     ErrIntOverflow
    0.53
    Activations Density 0.262%

    No Known Activations