© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 18-GEMMASCOPE-RES-16K
    11. 1726
    Prev
    Next
    INDEX
    Explanations

    mentions of specific individuals related to legal proceedings or cases

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

    Text after numbers

    np_acts-logits-general · gemini-2.0-flash

    Botswana, Gaborone, Lesotho

    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
     Gra
    -0.52
    #
    -0.51
     CreateTagHelper
    -0.47
     Hands
    -0.45
    []):
    -0.45
     Gran
    -0.45
    Gran
    -0.45
     gran
    -0.44
     Kula
    -0.43
    Hands
    -0.43
    POSITIVE LOGITS
     propOrder
    0.85
     Botswana
    0.83
     purpoſe
    0.81
     pleaſure
    0.78
     Gabor
    0.71
    ]--;
    0.70
     Mön
    0.70
     houſe
    0.69
     ſtate
    0.69
     leaſt
    0.69
    Activations Density 0.119%

    No Known Activations