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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 3335
    Prev
    Next
    INDEX
    Explanations

    emphasized concepts or critical points in discussions

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    ſicht
    -0.54
    ntable
    -0.52
    omitempty
    -0.51
    
    -0.51
    lalom
    -0.49
     Chwiliwch
    -0.49
    Caro
    -0.49
    \{\\
    -0.48
     ligiloj
    -0.48
    uity
    -0.48
    POSITIVE LOGITS
    devším
    0.56
    何より
    0.54
     surtout
    0.46
     alvor
    0.43
    重要的是
    0.42
     soprattutto
    0.42
     importantly
    0.40
     especially
    0.40
    secuencias
    0.40
     především
    0.39
    Activations Density 0.171%

    No Known Activations