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. 11-GEMMASCOPE-RES-16K
    6. 5813
    Prev
    Next
    INDEX
    Explanations

    phrases related to business interactions and customer relations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Hook Layer
    11
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    XmlAccessorType
    -0.71
     avoient
    -0.67
    脚注の使い方
    -0.66
    LookAnd
    -0.65
    حوالہ
    -0.64
     pouvoit
    -0.64
     Савезне
    -0.64
     auroit
    -0.63
    MessageTagHelper
    -0.62
     djur
    -0.62
    POSITIVE LOGITS
     reciprocal
    0.49
    tainer
    0.48
     reciproc
    0.47
    еме
    0.47
     recipro
    0.43
    rrggbb
    0.43
    BarItem
    0.40
    parency
    0.39
     mutual
    0.39
     friendship
    0.39
    Activations Density 0.401%

    No Known Activations