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. 17-GEMMASCOPE-RES-16K
    6. 407
    Prev
    Next
    INDEX
    Explanations

    mentions of player injuries or medical statuses related to athletes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 17-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_17/width_16k/average_l0_77
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.17.hook_resid_post
    Hook Layer
    17
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    </thead>
    -0.56
    excellents
    -0.48
     espèce
    -0.47
    quilibrium
    -0.46
     cached
    -0.45
     Corr
    -0.42
    forsch
    -0.42
    Corr
    -0.42
    bber
    -0.42
    verzeichnis
    -0.42
    POSITIVE LOGITS
     injury
    0.89
     injured
    0.83
     Injury
    0.80
    ModelSerializer
    0.79
     injuries
    0.77
     blessé
    0.77
     Injuries
    0.74
    injury
    0.73
    Injury
    0.72
     Injured
    0.71
    Activations Density 0.131%

    No Known Activations