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 - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 1452
    Prev
    Next
    INDEX
    Explanations

    references to specific historical regions and their relationships within the context of Schleswig and Holstein

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 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
     uſed
    -0.33
    Vlastnosti
    -0.32
     pép
    -0.29
     schimb
    -0.28
    \{\\
    -0.28
     eſſ
    -0.28
     anſ
    -0.27
     fufficient
    -0.27
     étudier
    -0.27
     crise
    -0.27
    POSITIVE LOGITS
    GOTREF
    0.65
     oprot
    0.64
     оригіналу
    0.63
    
    0.57
    FormTagHelper
    0.56
    FunctionFlags
    0.55
     Suits
    0.54
     Lokales
    0.54
    IsContent
    0.53
     kasarigan
    0.53
    Activations Density 1.039%

    No Known Activations