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. 77203
    Prev
    Next
    INDEX
    Explanations

    special characters or formatting elements in the text

    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
    Tikang
    -0.56
    hdashline
    -0.56
     tig
    -0.53
     rati
    -0.53
    LocationManager
    -0.53
     घट
    -0.53
     segment
    -0.52
     Ciro
    -0.51
    <td>
    -0.50
     conting
    -0.50
    POSITIVE LOGITS
     \
    0.91
    \{\\
    0.77
     împre
    0.66
     zijne
    0.62
     veiligheid
    0.62
     "\
    0.61
     femei
    0.60
     komende
    0.60
     gezondheid
    0.60
     către
    0.59
    Activations Density 0.117%

    No Known Activations