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. 9-GEMMASCOPE-RES-16K
    6. 6453
    Prev
    Next
    INDEX
    Explanations

    function calls and object-oriented programming constructs

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

    Code tokens after periods or arrows

    np_acts-logits-general · gemini-2.0-flash

    code followed by control flow statements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     noDo
    -0.48
    httphttps
    -0.47
    AutoField
    -0.46
    CardContent
    -0.45
    acular
    -0.45
    MessageTagHelper
    -0.44
     getRule
    -0.44
    TextWatcher
    -0.43
    house
    -0.43
    Izvori
    -0.43
    POSITIVE LOGITS
    (&:
    0.59
     Meksiku
    0.42
    ={({
    0.41
     gespeichert
    0.40
    seamnă
    0.39
    fromnode
    0.38
    ()?.
    0.37
    __.__
    0.36
    ുറ
    0.36
    >({
    0.35
    Activations Density 0.838%

    No Known Activations