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-2-9B
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 53730
    Prev
    Next
    INDEX
    Explanations

    programming language constructs and data structure definitions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     Whilst
    -0.48
     navidad
    -0.47
     whilst
    -0.46
    Whilst
    -0.45
     perchè
    -0.43
     jewellery
    -0.43
     dimana
    -0.42
    TypedDataSet
    -0.41
    為に
    -0.40
    chossen
    -0.40
    POSITIVE LOGITS
     poffible
    0.61
     Chriftian
    0.59
    ſelf
    0.58
     myſelf
    0.56
     Jefus
    0.56
     leaſt
    0.55
     fubject
    0.55
     Eſ
    0.54
     ſte
    0.53
     purpoſe
    0.53
    Activations Density 0.129%

    No Known Activations