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. 18-GEMMASCOPE-RES-16K
    6. 1003
    Prev
    Next
    INDEX
    Explanations

    phrases related to formal introductions and etiquette

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

    titles, names, or introductions

    np_acts-logits-general · gemini-2.0-flash

    honorifics and nicknames

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    BufferException
    -0.79
    aarrggbb
    -0.66
    ]))
    
    -0.65
     estekak
    -0.64
    balleur
    -0.62
    chism
    -0.61
    ########.
    -0.61
    ']))
    
    -0.59
    tegra
    -0.57
    ]));
    
    -0.57
    POSITIVE LOGITS
     title
    1.21
     titles
    1.19
     nickname
    1.10
     Titles
    1.09
     nicknames
    1.01
    title
    1.01
    Titles
    0.99
     Title
    0.97
     TITLE
    0.97
     surname
    0.96
    Activations Density 0.163%

    No Known Activations