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-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 3311
    Prev
    Next
    INDEX
    Explanations

    phrases about player injuries and their status in sports contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     NSCoder
    -0.37
     silencio
    -0.36
     Albert
    -0.33
     silêncio
    -0.31
     Aku
    -0.31
     concept
    -0.30
     Krause
    -0.30
     des
    -0.30
     comuniques
    -0.29
     cikk
    -0.29
    POSITIVE LOGITS
     resuming
    0.64
     resumption
    0.63
     reopening
    0.63
    windowFixed
    0.59
    adaptiveStyles
    0.58
     resumed
    0.57
     Normdatei
    0.57
     restored
    0.57
     reinstatement
    0.55
     reopened
    0.54
    Activations Density 0.155%

    No Known Activations