© 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. 20-GEMMASCOPE-RES-16K
    6. 5505
    Prev
    Next
    INDEX
    Explanations

    connections or transitions related to conditions and their impacts

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

    preceding "without"

    np_acts-logits-general · gemini-2.0-flash

    with or without

    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
    sprechend
    -0.58
    RectangleBorder
    -0.57
    archiviato
    -0.55
    modb
    -0.51
    GOTREF
    -0.50
    BackStack
    -0.49
    étisme
    -0.48
    новништво
    -0.48
    GIVEREF
    -0.48
    WriteTagHelper
    -0.48
    POSITIVE LOGITS
     without
    3.29
     Without
    3.01
    without
    2.95
    Without
    2.94
     WITHOUT
    2.60
     zonder
    2.51
     ohne
    2.47
    WITHOUT
    2.42
     tanpa
    2.39
     senza
    2.35
    Activations Density 0.818%

    No Known Activations