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. 2777
    Prev
    Next
    INDEX
    Explanations

    scientific paper titles and references

    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
     initComponents
    -1.06
    principalColumn
    -0.96
     kasarigan
    -0.89
    WireFormatLite
    -0.80
    ViewImports
    -0.76
    PyExc
    -0.75
    webElementXpaths
    -0.73
    GraphicsUnit
    -0.72
    verwijspagina
    -0.72
     puissiez
    -0.72
    POSITIVE LOGITS
     https
    0.59
     selfie
    0.58
     microbiome
    0.58
     selfies
    0.56
    https
    0.55
     ۲۰۱
    0.53
     arXiv
    0.53
     epigenetic
    0.52
     graphene
    0.52
     tweeted
    0.52
    Activations Density 0.804%

    No Known Activations