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. 14584
    Prev
    Next
    INDEX
    Explanations

    formatting elements indicative of section headers or document structure

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

    HTML heading tag: </h4>

    np_acts-logits-general · gemini-2.0-flash

    closing code tags and parentheses

    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
    LookAnd
    -0.54
     Plant
    -0.47
    Plant
    -0.46
     plant
    -0.46
     $#
    -0.45
    Chham
    -0.44
    html
    -0.44
    plant
    -0.43
    encodeWith
    -0.43
    EDEFAULT
    -0.43
    POSITIVE LOGITS
    ")));
    
    0.82
     referenties
    0.79
    RegressionTest
    0.79
    "]));
    0.76
    "])
    
    0.76
    '])
    
    0.74
    "){
    
    0.73
    ']))
    
    0.72
    ')));
    0.72
    ')}}"
    0.72
    Activations Density 0.017%

    No Known Activations