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. 20-GEMMASCOPE-RES-16K
    6. 5330
    Prev
    Next
    INDEX
    Explanations

    relationships and interdependencies among various systems and entities

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

    Things working or existing together

    np_acts-logits-general · gemini-2.0-flash

    togetherness

    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
     retrait
    -0.43
    SSM
    -0.43
    isPrime
    -0.42
    Eddy
    -0.41
    Chham
    -0.41
    cnia
    -0.41
     aqueles
    -0.41
    >):
    -0.40
    twice
    -0.40
     sowohl
    -0.39
    POSITIVE LOGITS
     together
    2.14
    together
    1.82
     juntos
    1.81
     Together
    1.78
    Together
    1.72
     TOGETHER
    1.69
     elkaar
    1.65
     saling
    1.56
     jointly
    1.55
     mutually
    1.50
    Activations Density 1.564%

    No Known Activations