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. 22-GEMMASCOPE-RES-16K
    6. 12903
    Prev
    Next
    INDEX
    Explanations

    terms related to variability and differences in data or conditions

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

    that relate to variance or difference

    np_acts-logits-general · gemini-2.0-flash

    vary by, between, with

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    testens
    -0.68
    Rhestr
    -0.60
     wikipagina
    -0.60
    كذا
    -0.54
    Pty
    -0.53
    chehen
    -0.52
     vermelhas
    -0.52
    ussian
    -0.51
    oulli
    -0.51
     plebis
    -0.51
    POSITIVE LOGITS
     widely
    0.89
     greatly
    0.83
     wildly
    0.79
     among
    0.66
     between
    0.64
    клопе
    0.62
     vastly
    0.61
     considerably
    0.61
     Widely
    0.60
     significantly
    0.59
    Activations Density 0.261%

    No Known Activations