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-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 91463
    Prev
    Next
    INDEX
    Explanations

    mentions of salads and salad-related dishes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    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
     kişinin
    -0.45
    AppMethodBeat
    -0.44
     Wey
    -0.44
    									
    -0.42
     kurie
    -0.42
    					
    -0.41
     krát
    -0.41
     persoane
    -0.41
     egens
    -0.40
     Cree
    -0.40
    POSITIVE LOGITS
     Salad
    1.41
     salad
    1.39
    Salad
    1.38
    salad
    1.26
     salads
    1.24
     salade
    0.96
    🥗
    0.89
     салат
    0.78
     Salat
    0.77
     InputDecoration
    0.74
    Activations Density 0.001%

    No Known Activations