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. 2449
    Prev
    Next
    INDEX
    Explanations

    references to comparisons or relationships involving the concept of "between."

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

    "between"

    np_acts-logits-general · gemini-2.0-flash

    introducing a range or interval

    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
    ly
    -0.64
    に
    -0.49
    erweise
    -0.48
    뮬
    -0.44
    AND
    -0.43
    otts
    -0.43
    تفصیلات
    -0.42
    and
    -0.41
     adaptées
    -0.41
     complètes
    -0.41
    POSITIVE LOGITS
     nahilalakip
    1.38
     betweenstory
    1.36
    complexContent
    1.17
     CreateTagHelper
    1.16
    WebVitals
    1.13
    enumi
    1.13
    tagHelperRunner
    1.10
     surla
    1.08
    protoimpl
    1.06
    enumii
    1.04
    Activations Density 0.131%

    No Known Activations