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. 10127
    Prev
    Next
    INDEX
    Explanations

    phrases or commands related to leaving or exiting situations

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

    Following "get"

    np_acts-logits-general · gemini-2.0-flash

    get/got + preposition/adverb

    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
     Durchsch
    -0.55
     élector
    -0.49
    enschein
    -0.47
    orologio
    -0.47
    wachung
    -0.46
    уме
    -0.45
     sanitarias
    -0.45
     bocas
    -0.45
     phẩm
    -0.45
    ❹
    -0.44
    POSITIVE LOGITS
     involved
    0.76
    IsContent
    0.74
    مصادر
    0.70
    SharedDtor
    0.68
     initComponents
    0.65
    Hochspringen
    0.64
     disambiguazione
    0.63
    addGap
    0.62
     betweenstory
    0.60
     nahilalakip
    0.60
    Activations Density 0.074%

    No Known Activations