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-2-2B
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 7082
    Prev
    Next
    INDEX
    Explanations

    names of individuals or figures, particularly from Islamic history or leadership

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    BeginContext
    -0.47
     sẻ
    -0.44
    ItemGroup
    -0.44
    ⓧ
    -0.44
     continúas
    -0.43
     desconoc
    -0.41
     BorderRadius
    -0.40
    laris
    -0.40
    propTypes
    -0.40
    Pare
    -0.39
    POSITIVE LOGITS
     يتيمه
    0.59
     <=",
    0.59
     uſed
    0.58
    InputBorder
    0.57
    soort
    0.55
     francès
    0.55
    ValueStyle
    0.53
    βολ
    0.53
    قایناقلار
    0.53
    tagHelper
    0.52
    Activations Density 0.108%

    No Known Activations