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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 131k
    5. 23-GEMMASCOPE-RES-131K
    6. 35
    Prev
    Next
    INDEX
    Explanations

    expressions of affection or emotional attachment

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_23/width_131k/average_l0_101
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.23.hook_resid_post
    Hook Layer
    23
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     незавершена
    -0.68
    AndEndTag
    -0.63
    WriteTagHelper
    -0.62
    Geplaatst
    -0.61
     >=",
    -0.60
    posedge
    -0.59
    encodeWith
    -0.59
     wikipagina
    -0.58
    MemoryWarning
    -0.58
    bitField
    -0.57
    POSITIVE LOGITS
    
    0.41
     okuyayım
    0.40
    is
    0.33
     ​​
    0.33
    KommentareTeilen
    0.32
    <blockquote>
    0.31
    ViewFeatures
    0.31
     don
    0.30
    [toxicity=0]
    0.30
      
    0.30
    Activations Density 0.003%

    No Known Activations