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. 1095
    Prev
    Next
    INDEX
    Explanations

    concepts related to moments and experiences that emphasize introspection and personal reflection

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

    references to the significance of small moments within a larger context

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

    daily activities

    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
    mmation
    -0.48
    zkod
    -0.48
    boots
    -0.47
    libft
    -0.46
    XtraEditors
    -0.46
     Brains
    -0.43
     مشين
    -0.43
    schul
    -0.42
    :],
    -0.42
     Yards
    -0.42
    POSITIVE LOGITS
     <<<<<<<<<<<<<<
    0.77
     things
    0.76
     essentials
    0.75
     daily
    0.74
     يتيمه
    0.74
    essentials
    0.73
     everyday
    0.72
     ordinary
    0.70
     AssemblyProduct
    0.69
     nahilalakip
    0.69
    Activations Density 0.074%

    No Known Activations