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-2-2B
    4. Residual Stream - 65k
    5. 16-GEMMASCOPE-RES-65K
    6. 52936
    Prev
    Next
    INDEX
    Explanations

    contractions and possessive forms

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

    possessive my or contractions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    +#+#
    -0.89
     BoxDecoration
    -0.86
     AssemblyCulture
    -0.85
     avoient
    -0.84
     pinulongan
    -0.82
     للاسماء
    -0.81
     ivelany
    -0.81
    TagMode
    -0.78
     InputDecoration
    -0.77
     étoient
    -0.76
    POSITIVE LOGITS
     I
    0.89
     my
    0.71
     myself
    0.62
     me
    0.57
     Mom
    0.55
    嫌
    0.54
     pretty
    0.53
     
    0.53
    </blockquote>
    0.52
    Anyways
    0.50
    Activations Density 0.073%

    No Known Activations