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
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 103217
    Prev
    Next
    INDEX
    Explanations

    specific programming language constructs and formatting related to functional requirements and validations in code

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

    Syntax elements and punctuation in programming code, especially those that terminate statements, define conditions, specify time units, or mark documentation boundaries.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
    Remote
    -0.45
    expect
    -0.41
    ametro
    -0.40
    remote
    -0.39
     Remote
    -0.39
     REMOTE
    -0.39
    laying
    -0.37
     ControllerBase
    -0.36
    eros
    -0.36
    зумі
    -0.36
    POSITIVE LOGITS
     noDo
    0.72
    BeginContext
    0.70
    IntoConstraints
    0.67
    发表于
    0.64
    Autoritní
    0.60
    HtmlAttribute
    0.59
    complexContent
    0.57
     Wicidata
    0.56
     Савезне
    0.55
    ngdoc
    0.55
    Activations Density 0.099%

    No Known Activations