Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 10-GEMMASCOPE-RES-16K
    6. 11613
    Prev
    Next
    INDEX
    Explanations

    structured data formats and elements of a web development context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Efq
    -0.83
     Jefus
    -0.81
     Monfieur
    -0.75
     becauſe
    -0.71
     quæ
    -0.70
    seamnă
    -0.70
    ſelf
    -0.69
    AndEndTag
    -0.69
     ]
    
    -0.68
     houſe
    -0.68
    POSITIVE LOGITS
     JADX
    0.56
    CodeAttribute
    0.49
    NewReader
    0.45
     AttributeSet
    0.45
    CLUS
    0.44
    onz
    0.44
    UrlResolution
    0.43
    SerializeField
    0.43
     BeautifulSoup
    0.42
     rowspan
    0.42
    Activations Density 0.152%

    No Known Activations