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 - 16k
    5. 11-GEMMASCOPE-RES-16K
    6. 5637
    Prev
    Next
    INDEX
    Explanations

    Java programming concepts and structures, particularly related to classes, methods, and data types

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Hook Layer
    11
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    存于互联网档案馆
    -0.77
     للمعارف
    -0.76
    FontOfSize
    -0.73
    tvguidetime
    -0.72
     متعلقه
    -0.71
     ſever
    -0.70
     myſelf
    -0.70
     NSCoder
    -0.68
     poffible
    -0.68
     ſche
    -0.67
    POSITIVE LOGITS
     my
    0.59
    tela
    0.58
    []
    0.57
    onedDateTime
    0.55
    IBOutlet
    0.54
     “
    0.53
     itu
    0.51
     start
    0.51
    HRESULT
    0.50
    Presence
    0.50
    Activations Density 0.236%

    No Known Activations