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-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 105127
    Prev
    Next
    INDEX
    Explanations

    code elements and structural components in a programming context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     endforeach
    -0.81
     Biôgrafia
    -0.77
    EndContext
    -0.77
     betweenstory
    -0.74
    存于互联网档案馆
    -0.69
    ſelf
    -0.68
     protoimpl
    -0.67
    artifactId
    -0.67
     ListTile
    -0.66
     nameof
    -0.65
    POSITIVE LOGITS
     the
    0.59
     pyta
    0.45
     probleem
    0.44
    if
    0.43
     metod
    0.42
    ommen
    0.42
     adecuadamente
    0.41
    STRUCTIONS
    0.41
     vrije
    0.41
     empêcher
    0.41
    Activations Density 0.055%

    No Known Activations