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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 7-GEMMASCOPE-RES-16K
    6. 4485
    Prev
    Next
    INDEX
    Explanations

    references to a root element or object in a structured data context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.7.hook_resid_post
    Hook Layer
    7
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    httphttps
    -0.84
    MIDDLEWARE
    -0.80
     Chamb
    -0.79
     Lari
    -0.74
    thodon
    -0.74
    żeli
    -0.72
     Belvedere
    -0.72
     SDC
    -0.71
    verwijspagina
    -0.70
     fibrillation
    -0.70
    POSITIVE LOGITS
     Root
    1.98
     root
    1.97
     ROOT
    1.91
     roots
    1.83
    ROOT
    1.81
    root
    1.80
     Roots
    1.77
    Root
    1.76
    Roots
    1.67
    roots
    1.67
    Activations Density 0.062%

    No Known Activations