© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 122645
    Prev
    Next
    INDEX
    Explanations

    references to design and development processes related to products or systems

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

    designed, built, or modeled

    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
    nezeu
    -0.46
     syn
    -0.36
    DIY
    -0.36
    Traducción
    -0.35
     DIY
    -0.35
     Hel
    -0.34
     кӀ
    -0.33
    BorderFactory
    -0.33
     Uk
    -0.33
    odend
    -0.32
    POSITIVE LOGITS
     informée
    0.59
    httphttps
    0.57
    onViewCreated
    0.52
    AddTagHelper
    0.51
     Empfang
    0.50
    étoit
    0.46
     Normdatei
    0.46
     Italij
    0.46
     ModelExpression
    0.46
     disambiguazione
    0.45
    Activations Density 0.118%

    No Known Activations