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. 1131
    Prev
    Next
    INDEX
    Explanations

    terms related to melee attacks and combat situations in close-range battles

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

    Preceding "is" or period (".")

    np_acts-logits-general · gemini-2.0-flash

    future action or description

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    httphttps
    -0.85
     kasarigan
    -0.80
     estekak
    -0.56
     utafitiHapana
    -0.56
     EconPapers
    -0.54
     CanadaChoose
    -0.54
     дописавши
    -0.53
     pinulongan
    -0.53
     للمعارف
    -0.52
     Савезне
    -0.52
    POSITIVE LOGITS
     occasion
    0.36
     orta
    0.34
     diversify
    0.33
     diversification
    0.33
     artık
    0.33
     ch
    0.32
     erstmals
    0.31
     occafion
    0.30
     alter
    0.29
     foray
    0.29
    Activations Density 0.448%

    No Known Activations