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-2-9B
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 5486
    Prev
    Next
    INDEX
    Explanations

    the word "before" and phrases indicating prior events or conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Gedicht
    -0.45
     Osun
    -0.45
    salms
    -0.43
    MLElement
    -0.43
    =$?
    -0.42
     Votes
    -0.40
     Wirkung
    -0.40
     arbete
    -0.40
    OECD
    -0.39
     Tatsache
    -0.39
    POSITIVE LOGITS
    before
    1.34
     before
    1.34
     Before
    1.19
    Before
    1.18
     BEFORE
    1.16
    BEFORE
    1.15
    efore
    0.94
     sebelum
    0.91
     før
    0.90
     antes
    0.87
    Activations Density 0.076%

    No Known Activations