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 - 65k
    5. 17-GEMMASCOPE-RES-65K
    6. 28418
    Prev
    Next
    INDEX
    Explanations

    the presence of authorship or citation in a document

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 17-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.17.hook_resid_post
    Hook Layer
    17
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    InjectMocks
    -0.81
     Exacts
    -0.75
    nhung
    -0.72
    Xaml
    -0.71
    ็จ
    -0.70
    childs
    -0.69
    rój
    -0.68
    antd
    -0.66
    gulier
    -0.66
    ühungen
    -0.65
    POSITIVE LOGITS
    s
    1.66
    }`
    0.84
    t
    0.78
    }'
    0.77
     ")";
    0.76
     "]";
    0.75
     "}";
    0.73
    €™
    0.72
    soever
    0.72
     Compagn
    0.72
    Activations Density 0.025%

    No Known Activations