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. 21-GEMMASCOPE-RES-16K
    6. 322
    Prev
    Next
    INDEX
    Explanations

    elements related to programming and game mechanics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     censiti
    -0.59
    inerja
    -0.58
     jealousy
    -0.58
     envy
    -0.58
    ніципа
    -0.57
     ampoule
    -0.57
    AddTagHelper
    -0.57
     spedizione
    -0.56
     cuckoo
    -0.56
     jealous
    -0.56
    POSITIVE LOGITS
     input
    1.46
    input
    1.33
    Input
    1.19
     Input
    1.13
     prompt
    1.13
    输入
    1.04
    prompt
    1.04
    INPUT
    1.03
    Prompt
    1.02
     inputs
    1.02
    Activations Density 0.184%

    No Known Activations