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. 999
    Prev
    Next
    INDEX
    Explanations

    phrases related to health, wellness, and technology services

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

    information, advice, or services

    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
    ロウィン
    -0.76
     дописавши
    -0.72
    expandindo
    -0.70
     resourceCulture
    -0.70
    enterOuterAlt
    -0.69
    ஢
    -0.69
    ஫
    -0.67
    jsxFileName
    -0.66
    +#+#
    -0.66
     يتيمه
    -0.66
    POSITIVE LOGITS
     eletrônico
    0.31
     élé
    0.26
     bodas
    0.26
     indipendente
    0.25
     abandonné
    0.25
     saurait
    0.24
    zunehmen
    0.24
     eletr
    0.24
     elettron
    0.23
     economía
    0.23
    Activations Density 0.037%

    No Known Activations