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. 10791
    Prev
    Next
    INDEX
    Explanations

    phrases related to public health studies and methodologies

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

    Follows "with", "selected", "representative", or "fact"

    np_acts-logits-general · gemini-2.0-flash

    sampling methods

    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
     &___
    -0.54
     (°
    -0.47
     Dickson
    -0.47
    zkod
    -0.46
    Wikis
    -0.46
     spek
    -0.46
    colade
    -0.46
     useStyles
    -0.46
     whis
    -0.44
     Proving
    -0.44
    POSITIVE LOGITS
     sample
    1.31
     sampling
    1.29
    sample
    1.25
     Sampling
    1.20
    sampling
    1.18
    Sample
    1.17
    Sampling
    1.16
     Sample
    1.16
     SAMPLE
    1.04
     sampled
    1.00
    Activations Density 0.762%

    No Known Activations