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. 6669
    Prev
    Next
    INDEX
    Explanations

    important information related to self-storage facilities and services

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

    after imperative verbs

    np_acts-logits-general · gemini-2.0-flash

    choosing or selecting options

    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
    المشاركات
    -0.50
    <?
    -0.49
     Tato
    -0.47
     provoc
    -0.45
    Portale
    -0.44
    emény
    -0.44
     incidente
    -0.43
    Xaml
    -0.43
    reactstrap
    -0.43
     ICD
    -0.42
    POSITIVE LOGITS
    Choosing
    1.35
     Choosing
    1.32
     choosing
    1.27
    choosing
    1.22
     selecting
    1.19
     Selecting
    1.14
    Selecting
    1.12
     selection
    1.06
    選ぶ
    1.04
     choose
    1.03
    Activations Density 0.203%

    No Known Activations