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
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 7345
    Prev
    Next
    INDEX
    Explanations

    structured programming elements and commands, particularly in codebases

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
    principalColumn
    -1.17
    fromnode
    -1.13
    ConstraintMaker
    -0.93
    EDEFAULT
    -0.88
     للاسماء
    -0.85
    +#+#
    -0.84
    httphttps
    -0.83
     HasFactory
    -0.81
    PerformLayout
    -0.79
    <unused16>
    -0.78
    POSITIVE LOGITS
     LIRE
    0.21
     genere
    0.19
    Đ
    0.17
     Cité
    0.17
    ത്
    0.16
     bn
    0.15
    ollectionView
    0.15
    np
    0.15
     fubject
    0.15
    <em>
    0.15
    Activations Density 0.072%

    No Known Activations