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 - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 7180
    Prev
    Next
    INDEX
    Explanations

    file paths and directory structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Lycka
    -0.65
    tagext
    -0.65
    LookAnd
    -0.61
    GenerationType
    -0.59
     StringTokenizer
    -0.59
    MessageTagHelper
    -0.58
    NOPQRST
    -0.57
    maketitle
    -0.56
    StoreMessageInfo
    -0.55
     nemlig
    -0.55
    POSITIVE LOGITS
    Documents
    0.81
    Desktop
    0.77
    Dropbox
    0.73
    projects
    0.68
     projects
    0.68
    workspace
    0.68
     Projects
    0.67
    Users
    0.66
    Projects
    0.65
    Workspace
    0.65
    Activations Density 0.110%

    No Known Activations