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. 56541
    Prev
    Next
    INDEX
    Explanations

    references to legal processes and courtroom terminology

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

    legal, technical, and foreign terms

    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-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
     pinulongan
    -0.61
     propOrder
    -0.48
    Cursors
    -0.47
     '\\;'
    -0.46
     gron
    -0.46
    smithy
    -0.45
     Organis
    -0.45
     Silas
    -0.44
    IsMutable
    -0.44
    roda
    -0.44
    POSITIVE LOGITS
    MockBean
    0.46
     يتيمه
    0.43
    corazón
    0.42
     makl
    0.40
    pulseira
    0.39
     afield
    0.38
    jsonwebtoken
    0.38
     ellers
    0.38
    encodeWith
    0.38
     Gesang
    0.38
    Activations Density 0.000%

    No Known Activations