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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 69319
    Prev
    Next
    INDEX
    Explanations

    references to physical gaps or cracks in construction or materials

    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-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    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
    StoryboardSegue
    -0.60
    AppMethodBeat
    -0.53
    posedge
    -0.46
    SpringRunner
    -0.43
    RTSN
    -0.43
    addLast
    -0.42
    Pautan
    -0.41
    cuba
    -0.41
    hib
    -0.40
    benen
    -0.40
    POSITIVE LOGITS
     gaps
    0.56
     cracks
    0.53
     joints
    0.50
    gaps
    0.47
     Gaps
    0.47
     Cracks
    0.47
    Filler
    0.46
    windowFixed
    0.45
     fissures
    0.45
     leakage
    0.43
    Activations Density 0.013%

    No Known Activations