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

    Neuronpedia

    APICircuit 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. 20-GEMMASCOPE-RES-16K
    6. 10670
    Prev
    Next
    INDEX
    Explanations

    information related to Pakistani politicians and their political activities

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

    Locations or entities within Pakistan

    np_acts-logits-general · gemini-2.0-flash

    Pakistani geography

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    CppCodeGen
    -0.56
     Fairfield
    -0.55
     Koss
    -0.44
    Vee
    -0.44
    bent
    -0.42
     ethers
    -0.41
    toggleClass
    -0.41
     bending
    -0.40
    setImageBitmap
    -0.40
     Gerais
    -0.40
    POSITIVE LOGITS
     Pakistan
    1.07
     Islamabad
    0.97
     Lahore
    0.95
    Pakistan
    0.93
     Karachi
    0.93
     Pakistani
    0.89
    lamabad
    0.88
     Sindh
    0.87
     Baloch
    0.86
    WithIOException
    0.83
    Activations Density 0.082%

    No Known Activations