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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 20-GEMMASCOPE-RES-65K
    6. 595
    Prev
    Next
    INDEX
    Explanations

    mathematical symbols and expressions related to functions and their properties

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

    mathematical expressions and statistical notations

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

    mathematical notation including square roots

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
    Passive
    -0.51
    SequentialGroup
    -0.48
     domo
    -0.47
     onlyOwner
    -0.45
     passive
    -0.45
     autorytatywna
    -0.45
    CADE
    -0.44
     Passive
    -0.44
     trebui
    -0.44
     回
    -0.44
    POSITIVE LOGITS
    sqrt
    1.79
     sqrt
    1.68
     √
    1.59
    √
    1.57
     root
    1.52
    Sqrt
    1.41
     Root
    1.40
     racine
    1.37
    root
    1.37
    Root
    1.32
    Activations Density 1.009%

    No Known Activations