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. 20-GEMMASCOPE-RES-65K
    6. 1035
    Prev
    Next
    INDEX
    Explanations

    concepts related to the differences between real and virtual entities

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

    concepts related to virtual and real edges in directed graphical models

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

    real and currency

    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
    +#+#
    -0.88
     Harrow
    -0.67
     pinulongan
    -0.66
     Shakspeare
    -0.65
    +:+
    -0.64
     oprot
    -0.64
    #+#
    -0.64
    fromnode
    -0.64
     Corinne
    -0.62
     Benton
    -0.62
    POSITIVE LOGITS
    real
    2.29
     Real
    2.28
     real
    2.25
     REAL
    2.21
    Real
    2.19
    REAL
    2.16
    getReal
    1.79
     realism
    1.64
     realist
    1.56
     realis
    1.55
    Activations Density 0.040%

    No Known Activations