Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 20-GEMMASCOPE-RES-16K
    6. 2227
    Prev
    Next
    INDEX
    Explanations

    mentions of biological samples and data collection processes

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

    Collection of biological samples

    np_acts-logits-general · gemini-2.0-flash

    collecting medical samples

    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
     GenerationType
    -0.69
     BoxFit
    -0.66
    GenerationType
    -0.65
    DataAnnotations
    -0.64
    IUrlHelper
    -0.63
     تضيفلها
    -0.62
     ujednoznacz
    -0.61
    UnusedPrivate
    -0.61
    Enllaços
    -0.58
     täh
    -0.57
    POSITIVE LOGITS
     samples
    2.53
     sample
    2.35
     Samples
    2.15
    samples
    2.01
    sample
    2.00
    Samples
    1.98
     Sample
    1.96
    Sample
    1.92
     sampling
    1.90
     sampled
    1.84
    Activations Density 1.363%

    No Known Activations