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. 22-GEMMASCOPE-RES-16K
    6. 13847
    Prev
    Next
    INDEX
    Explanations

    references to archaeological findings and tools

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Ancient history or prehistoric times

    np_acts-logits-general · gemini-2.0-flash

    ancient human history and archaeology

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Restoration
    -0.48
     disposto
    -0.48
     mercy
    -0.47
    SuppressLint
    -0.46
    DataMember
    -0.46
     Freire
    -0.46
    FILTER
    -0.45
     libéral
    -0.45
    wyk
    -0.45
    nodoc
    -0.45
    POSITIVE LOGITS
     Neolithic
    0.82
     <>",
    0.78
     hunter
    0.74
     Stone
    0.66
     Bronze
    0.65
    UrlResolution
    0.65
    +#+#
    0.65
     prehistoric
    0.64
    はじめに
    0.64
     domestic
    0.64
    Activations Density 0.318%

    No Known Activations