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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    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. 812
    Prev
    Next
    INDEX
    Explanations

    elements of data structures or programming elements

    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
     Chriftian
    -1.04
     chofe
    -0.98
     fubject
    -0.97
     Monfieur
    -0.97
     myſelf
    -0.97
     itſelf
    -0.96
     Jefus
    -0.96
     Theſe
    -0.95
     themſelves
    -0.94
     himſelf
    -0.93
    POSITIVE LOGITS
    numerusform
    0.59
    principalColumn
    0.44
    tonode
    0.43
    ErrUnexpectedEOF
    0.43
    !*\
    0.42
     समीक्षक
    0.41
     crossorigin
    0.41
    MessageTagHelper
    0.40
    mergeFrom
    0.40
    <eos>
    0.40
    Activations Density 0.667%

    No Known Activations