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. 10332
    Prev
    Next
    INDEX
    Explanations

    references to patient populations and their potential benefits from medical treatments or therapies

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

    preceding "benefit" or similar words

    np_acts-logits-general · gemini-2.0-flash

    who might benefit

    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
     Inten
    -0.49
    utives
    -0.47
    __':
    -0.46
    erschiedene
    -0.45
    estrian
    -0.44
    amarca
    -0.44
     رت
    -0.44
    orexia
    -0.42
    ställ
    -0.42
    convertView
    -0.42
    POSITIVE LOGITS
     benefit
    2.58
     benefited
    2.45
    benefit
    2.30
     benefitted
    2.28
     benefiting
    2.26
     Benefit
    2.25
    Benefit
    2.20
     benefits
    2.01
     BENE
    1.97
    benefits
    1.74
    Activations Density 0.433%

    No Known Activations