© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 10590
    Prev
    Next
    INDEX
    Explanations

    numerical data and references to quantities in scientific or medical contexts

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

    Quantities or proportions

    np_acts-logits-general · gemini-2.0-flash

    quantities or fractions expressed with approximations

    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
    WriteTagHelper
    -0.68
    ENSE
    -0.50
    BeginContext
    -0.49
     createContext
    -0.48
    ceğim
    -0.48
    TemporalType
    -0.48
    IContainer
    -0.47
     betweenstory
    -0.47
    StackNavigator
    -0.47
    rspec
    -0.46
    POSITIVE LOGITS
     half
    3.49
    half
    3.00
    Half
    2.88
     Half
    2.81
     HALF
    2.65
    HALF
    2.40
     halves
    2.13
     metade
    2.13
     halv
    2.11
     helft
    2.10
    Activations Density 0.463%

    No Known Activations