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. 18-GEMMASCOPE-RES-16K
    6. 8756
    Prev
    Next
    INDEX
    Explanations

    references to specific locations, events, or casualties related to military actions

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

    Afghanistan and surrounding regions

    np_acts-logits-general · gemini-2.0-flash

    Afghanistan and Afghan people

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Lagos
    -0.51
    Lagos
    -0.50
     make
    -0.50
     makes
    -0.49
     ce
    -0.47
     Brésil
    -0.46
     Igbo
    -0.45
    Fos
    -0.44
     einsch
    -0.44
     nen
    -0.43
    POSITIVE LOGITS
     Afghanistan
    1.34
     Afghan
    1.25
     Kabul
    1.25
     Taliban
    1.22
    Afghanistan
    1.20
    Afghan
    1.15
     afghan
    1.12
     Afghans
    1.11
     Afgan
    1.10
     kasarigan
    0.97
    Activations Density 0.089%

    No Known Activations