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-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 5542
    Prev
    Next
    INDEX
    Explanations

    names and references related to Pakistani politics and politicians

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

    Pakistan, Pakistani

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     itſelf
    -0.60
     لئے
    -0.55
    HostException
    -0.54
    exitRule
    -0.53
    Rptr
    -0.53
    UnknownFields
    -0.51
     препратки
    -0.50
    NameInMap
    -0.50
    RegistryLite
    -0.49
     Etrus
    -0.49
    POSITIVE LOGITS
     Pakistan
    0.80
     Pakistani
    0.73
    Pakistan
    0.68
    lamabad
    0.61
     pakistan
    0.60
     Islamabad
    0.59
     Karachi
    0.48
     Lahore
    0.46
     Peshawar
    0.44
     Urdu
    0.41
    Activations Density 0.011%

    No Known Activations