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. 2-GEMMASCOPE-RES-16K
    6. 3758
    Prev
    Next
    INDEX
    Explanations

    instances of the word "scheme" and its variations, indicating a focus on deceptive or fraudulent plans

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 2-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    loaf
    -0.59
    fim
    -0.55
     Batu
    -0.54
    cub
    -0.52
     Otis
    -0.51
     Terms
    -0.49
     Bieber
    -0.48
     Gré
    -0.47
     PAR
    -0.47
     Raton
    -0.47
    POSITIVE LOGITS
     SCHEME
    1.46
     Schemes
    1.38
    schemes
    1.37
    scheme
    1.34
     scheme
    1.28
     Scheme
    1.27
     schemes
    1.25
    cheme
    1.17
    Schemes
    1.14
    Scheme
    1.11
    Activations Density 0.052%

    No Known Activations