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. 45732
    Prev
    Next
    INDEX
    Explanations

    legal terminology and references within a judicial context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    mandamus

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    GraphicsUnit
    -0.78
    /*
    -0.74
    __":
    
    -0.70
    __':
    
    -0.69
    
    -0.66
    __":
    -0.66
    AddTagHelper
    -0.66
    +:+
    -0.66
     BoxFit
    -0.65
     snippetHide
    -0.65
    POSITIVE LOGITS
    iramente
    0.31
    pecabezas
    0.29
     Pueden
    0.28
     saludos
    0.27
     Bühne
    0.27
     Speicher
    0.26
     하는
    0.26
     tahap
    0.26
     Punkt
    0.25
     hebat
    0.24
    Activations Density 0.006%

    No Known Activations