Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlog/PodcastSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 2036
    Prev
    Next
    INDEX
    Explanations

    technical terms related to imaging and feature extraction in neural networks

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 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
     fidanz
    -0.49
    Indented
    -0.48
     Dib
    -0.47
    ഖ
    -0.47
     photographed
    -0.46
    柱
    -0.46
     National
    -0.46
     upgrading
    -0.45
    Slug
    -0.44
     cutters
    -0.44
    POSITIVE LOGITS
     vettoriale
    0.85
     computationally
    0.84
     vectoriales
    0.78
     vetorial
    0.76
     esternos
    0.75
    polazione
    0.73
     wikipagina
    0.72
     artificiales
    0.72
     geométrico
    0.72
     químicas
    0.72
    Activations Density 3.231%

    No Known Activations