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. 6630
    Prev
    Next
    INDEX
    Explanations

    phrases that refer to inclusivity and the presence of multiple elements or features

    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
    and
    -0.54
     scot
    -0.49
    that
    -0.49
    C
    -0.48
    tro
    -0.47
    ally
    -0.47
    which
    -0.46
    baomidou
    -0.46
    retario
    -0.45
    instead
    -0.45
    POSITIVE LOGITS
     includes
    2.82
     include
    2.49
     Includes
    2.47
    Includes
    2.34
    includes
    2.30
     INCLUDES
    2.10
     inclui
    1.99
     Include
    1.99
     incluye
    1.99
     INCLUDE
    1.81
    Activations Density 0.171%

    No Known Activations