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. 4-GEMMASCOPE-RES-16K
    6. 3383
    Prev
    Next
    INDEX
    Explanations

    terms and phrases related to chemistry, specifically those that describe substances, chemicals, and their characteristics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.4.hook_resid_post
    Hook Layer
    4
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    AutoScale
    -0.63
    )}+
    -0.61
     Betten
    -0.56
     intptr
    -0.53
     île
    -0.52
    Voted
    -0.51
     défis
    -0.50
     समीक्षक
    -0.49
     chi̍t
    -0.48
    /*",
    -0.47
    POSITIVE LOGITS
    													
    1.08
    														
    1.07
    															
    1.05
    											
    0.99
    																
    0.93
    												
    0.92
    										
    0.91
    UrlResolution
    0.87
    																	
    0.86
    MethodManager
    0.83
    Activations Density 0.458%

    No Known Activations