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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 441
    Prev
    Next
    INDEX
    Explanations

    elements and structures related to data processing and manipulation in programming contexts

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

    programming syntax and code-related tokens, particularly involving pairs, updates, and data structures.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    __(/*!
    -0.52
    MessageTagHelper
    -0.51
    tonode
    -0.50
     оригіналу
    -0.49
    loroethene
    -0.48
    पया
    -0.47
    rrggbb
    -0.47
     insuffisamment
    -0.45
    lapsingToolbar
    -0.45
    IntoConstraints
    -0.44
    POSITIVE LOGITS
     pair
    0.63
     pairs
    0.59
     paire
    0.58
     couples
    0.56
    ArgsConstructor
    0.53
     couple
    0.53
     Pair
    0.52
     duo
    0.52
     coppia
    0.52
    pair
    0.51
    Activations Density 0.520%

    No Known Activations