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. 30456
    Prev
    Next
    INDEX
    Explanations

    complex mathematical expressions or equations involving roots and their properties

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

    mathematical and code structures

    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
     queſta
    -1.34
    ロウィン
    -1.27
    ſchaft
    -1.25
    ſicht
    -1.24
    niſſe
    -1.23
    ſſung
    -1.21
    <unused14>
    -1.20
    <unused16>
    -1.20
    <unused8>
    -1.20
    [@BOS@]
    -1.20
    POSITIVE LOGITS
    I
    0.56
        
    0.55
    	
    0.55
    _
    0.55
    (
    0.54
    The
    0.52
                
    0.50
            
    0.49
    2
    0.49
                    
    0.48
    Activations Density 0.026%

    No Known Activations