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. 14649
    Prev
    Next
    INDEX
    Explanations

    scientific terms and technical details related to various fields

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

    following prepositions or symbols

    np_acts-logits-general · gemini-2.0-flash

    chemical shifts of Te and Bi

    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
    </h3>
    -0.87
    >';
    
    -0.80
    </em>
    -0.79
    ';
    
    -0.76
    '
    
    -0.74
    '),
    
    -0.72
    tvguidetime
    -0.69
    ')));
    -0.65
    '])){
    
    -0.65
    ':
    
    -0.64
    POSITIVE LOGITS
    AddTagHelper
    0.64
    numerusform
    0.59
    <bos>
    0.52
    ScopeManager
    0.49
    forChild
    0.47
    ”).
    0.46
    vail
    0.46
    évaluateur
    0.44
    .”.
    0.44
    </u>
    0.43
    Activations Density 2.149%

    No Known Activations