Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 105143
    Prev
    Next
    INDEX
    Explanations

    code structure and elements related to functions and data processing

    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
    Autoritní
    -0.34
    preso
    -0.30
     quoque
    -0.29
    ほか
    -0.29
     atque
    -0.28
    Coh
    -0.27
     Nahrungs
    -0.26
     miten
    -0.26
    counter
    -0.26
    RECIP
    -0.26
    POSITIVE LOGITS
    RenderAtEndOf
    0.65
     yyhl
    0.63
     незавершена
    0.57
    InlineData
    0.56
    
    0.55
    pausal
    0.55
     "..\..\..\
    0.55
    osexuality
    0.54
    SBATCH
    0.53
    $_['
    0.53
    Activations Density 0.015%

    No Known Activations