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. 15-GEMMASCOPE-RES-16K
    6. 8711
    Prev
    Next
    INDEX
    Explanations

    HTML tags and structure related to web development documents

    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
     архивлан
    -1.08
     Италијани
    -1.04
    oredCriteria
    -1.03
    WithIOException
    -1.01
     myſelf
    -0.97
    WireFormatLite
    -0.96
    webElementXpaths
    -0.96
     ivelany
    -0.94
     esternos
    -0.90
     himſelf
    -0.90
    POSITIVE LOGITS
    =<?
    0.46
    äls
    0.43
     mon
    0.38
    óc
    0.37
     Pol
    0.37
    Dispose
    0.37
    ]<=
    0.36
    Mon
    0.36
     nar
    0.35
     Mon
    0.35
    Activations Density 0.180%

    No Known Activations