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 - 65k
    5. 16-GEMMASCOPE-RES-65K
    6. 53074
    Prev
    Next
    INDEX
    Explanations

    quotes and statements from notable authors

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

    quotes and sayings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     المعيارى
    -0.59
    Clik
    -0.55
    脚注の使い方
    -0.55
     незавершена
    -0.54
    gdx
    -0.54
    Xna
    -0.53
    引き続き
    -0.52
    ArgumentParser
    -0.50
    ącz
    -0.49
    __(/*!
    -0.49
    POSITIVE LOGITS
    LookAnd
    0.70
     Италијани
    0.62
     Theſe
    0.60
    msgSender
    0.56
    iaomi
    0.56
    Quotes
    0.54
    ipot
    0.54
    JpaRepository
    0.52
    BufferException
    0.52
     Tzu
    0.52
    Activations Density 0.029%

    No Known Activations