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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 8597
    Prev
    Next
    INDEX
    Explanations

    medical and legal terminology related to injury or impairment

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

    Following tokens like "in", "ex", "cent", "pu"

    np_acts-logits-general · gemini-2.0-flash

    forming words or phrases

    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
     “
    -0.80
    ]));
    
    -0.78
    -
    -0.78
    ()));
    
    -0.69
    <eos>
    -0.67
    ())));
    -0.66
    }));
    
    -0.66
    /
    -0.66
     «
    -0.64
    man
    -0.64
    POSITIVE LOGITS
     ujednoznacz
    1.12
     مشين
    1.03
     المعيارى
    1.01
    CloseOperation
    0.97
     varandra
    0.96
     détru
    0.94
    存于互联网档案馆
    0.94
     ainfi
    0.92
    IntoConstraints
    0.91
     Audiodateien
    0.91
    Activations Density 0.442%

    No Known Activations