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. 68466
    Prev
    Next
    INDEX
    Explanations

    mentions of scoring and points in sports contexts

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

    leg kick or penalty

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    SBATCH
    -0.52
    featureID
    -0.50
    dient
    -0.50
    ſelf
    -0.49
    WebElementEntity
    -0.49
    ########.
    -0.48
    FullName
    -0.47
    didSet
    -0.47
     ſehr
    -0.47
     onBackPressed
    -0.46
    POSITIVE LOGITS
     penalty
    0.44
    PerformLayout
    0.43
    labelControl
    0.39
     penalties
    0.35
    tonsoft
    0.34
    penalty
    0.34
     Penalty
    0.33
     BufferedWriter
    0.33
    RetentionPolicy
    0.32
    borderFill
    0.31
    Activations Density 0.002%

    No Known Activations