Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 9736
    Prev
    Next
    INDEX
    Explanations

    phrases indicating opinions or evaluations of experiences, particularly in the context of entertainment or products

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
     unbelievably
    -1.22
     absolutely
    -1.21
     incredibly
    -1.16
     utterly
    -1.13
    absolutely
    -1.11
     Absolutely
    -1.08
     absolument
    -1.07
     incredible
    -1.05
     incroyable
    -1.04
     unbelievable
    -1.03
    POSITIVE LOGITS
     nice
    1.13
     decent
    0.99
     pleasant
    0.98
    nice
    0.95
     enjoyable
    0.94
     good
    0.94
     interesting
    0.93
     useful
    0.90
     Nice
    0.88
     agreeable
    0.88
    Activations Density 0.521%

    No Known Activations