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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 9-GEMMASCOPE-RES-16K
    6. 15111
    Prev
    Next
    INDEX
    Explanations

    the term "ignore" in various contexts, indicating a focus on documentation or instructions regarding items to be omitted or disregarded

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 9-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Ruh
    -0.84
    k
    -0.84
     boissons
    -0.78
    ب
    -0.77
     EDS
    -0.75
     titolata
    -0.73
     états
    -0.72
     piedi
    -0.71
    udah
    -0.70
     colpa
    -0.70
    POSITIVE LOGITS
    ]--;
    0.89
    íso
    0.83
    bootstrapcdn
    0.77
     Exor
    0.73
    etcode
    0.73
    lloworld
    0.72
    NORE
    0.72
     Helper
    0.71
    ffix
    0.70
    icestershire
    0.69
    Activations Density 0.142%

    No Known Activations