Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 15-GEMMASCOPE-RES-16K
    6. 2754
    Prev
    Next
    INDEX
    Explanations

    keywords used in programming languages, particularly related to functions and data manipulation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    TextBoxColumn
    -0.51
    ⓧ
    -0.50
    ISupport
    -0.49
     numberWith
    -0.49
     Vikipedi
    -0.47
    JNIEnv
    -0.46
    etero
    -0.45
    Vidite
    -0.45
     numbers
    -0.44
    いくら
    -0.44
    POSITIVE LOGITS
    hyrchwyd
    0.84
     wireType
    0.69
     mapStateToProps
    0.64
    webElementXpaths
    0.62
     Scenes
    0.60
     BorderSide
    0.59
    CloseOperation
    0.56
    bitField
    0.55
    gebob
    0.54
     jspb
    0.54
    Activations Density 0.237%

    No Known Activations