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. 16-GEMMASCOPE-RES-65K
    6. 53105
    Prev
    Next
    INDEX
    Explanations

    the word "finally" or related expressions indicating a conclusion or finality

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    expressing conclusion or resolution

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     OMITTED
    -0.67
     znam
    -0.57
    ğim
    -0.57
     verhält
    -0.57
    charCodeAt
    -0.56
    COCK
    -0.56
     Saks
    -0.55
    σσ
    -0.54
     #%
    -0.54
     objectMapper
    -0.54
    POSITIVE LOGITS
    GOTREF
    0.79
    íncia
    0.73
    ientôt
    0.71
    arrhea
    0.70
    jsdelivr
    0.69
     Goh
    0.68
    ValueStyle
    0.67
    PerformLayout
    0.66
    ंदीखरीदारी
    0.66
     bref
    0.66
    Activations Density 0.012%

    No Known Activations