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-2-2B
    4. Residual Stream - 65k
    5. 17-GEMMASCOPE-RES-65K
    6. 28553
    Prev
    Next
    INDEX
    Explanations

    expressions of customer satisfaction and comfort during an experience

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 17-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.17.hook_resid_post
    Hook Layer
    17
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     relatively
    -0.70
    比較的
    -0.69
    Generally
    -0.68
     agak
    -0.66
     Generally
    -0.66
    relatively
    -0.63
     Relatively
    -0.63
     moderately
    -0.62
     relativamente
    -0.61
     umumnya
    -0.61
    POSITIVE LOGITS
     unbelievable
    0.83
    Incredible
    0.79
     incredible
    0.77
     unimaginable
    0.77
    incredible
    0.76
     unglaub
    0.75
     unforgettable
    0.74
     indescribable
    0.73
     unbelievably
    0.72
     unspeakable
    0.72
    Activations Density 0.008%

    No Known Activations