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-2-9B
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 62305
    Prev
    Next
    INDEX
    Explanations

    references to personal experiences or preferences

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

    Connecting phrases or transitions that link ideas, indicate decisions, or show the progression of thought in text.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    =?,
    -0.33
    Rujuakan
    -0.33
    ✭✭
    -0.32
    URBANA
    -0.30
     flavors
    -0.30
     osv
    -0.30
     betrug
    -0.29
    Cb
    -0.28
    طق
    -0.28
    こと
    -0.28
    POSITIVE LOGITS
    RenderAtEndOf
    0.65
     nonUne
    0.58
     <<<<<<<<<<<<<<
    0.54
    SourceChecksum
    0.52
     ModelRenderer
    0.52
     propOrder
    0.52
     aDecoder
    0.51
     CreateTagHelper
    0.50
    parsedMessage
    0.50
    WithIOException
    0.50
    Activations Density 0.042%

    No Known Activations