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-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 10233
    Prev
    Next
    INDEX
    Explanations

    syntax structures and function definitions in programming code

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     &&
    -0.52
    elif
    -0.48
     solange
    -0.43
     endwhile
    -0.43
    只要
    -0.42
    strcmp
    -0.41
    .
    -0.41
    for
    -0.41
    Optional
    -0.41
    TryGetValue
    -0.40
    POSITIVE LOGITS
     صوتيه
    0.89
     betweenstory
    0.86
     propOrder
    0.83
    脚注の使い方
    0.83
     pageContext
    0.81
    MemoryWarning
    0.81
    FieldBuilder
    0.79
     kasarigan
    0.77
    存于互联网档案馆
    0.74
    WritableDatabase
    0.74
    Activations Density 0.072%

    No Known Activations