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. 20-GEMMASCOPE-RES-16K
    6. 8
    Prev
    Next
    INDEX
    Explanations

    Java code constructs, particularly new object instantiations and list/collection declarations.

    oai_token-act-pair · claude-3-5-sonnet-20240620Triggered by @jbloomaus

    Java programming code involving data structures and method calls.

    oai_token-act-pair · gpt-3.5-turboTriggered by @jbloomaus

    programming-related constructs and structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 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
    AddTagHelper
    -0.75
    fromCharCode
    -0.63
    таратура
    -0.58
     Roskov
    -0.55
    rungsseite
    -0.55
     arşivlendi
    -0.54
     שוליים
    -0.51
     ActivatedRoute
    -0.51
     userSchema
    -0.50
    vierno
    -0.50
    POSITIVE LOGITS
     list
    2.66
     List
    2.56
    List
    2.52
    list
    2.51
     LIST
    2.40
    LIST
    2.29
     lists
    2.13
     Lists
    2.02
     liste
    1.97
     listing
    1.93
    Activations Density 0.154%

    No Known Activations