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. 53188
    Prev
    Next
    INDEX
    Explanations

    function calls and structure in a programming context

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

    asynchronous calls and callbacks

    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
    tanleria
    -0.76
     CreateTagHelper
    -0.69
     للمعارف
    -0.67
    StructEnd
    -0.65
     cherchés
    -0.65
     AssemblyCulture
    -0.63
    Datuak
    -0.62
     CURIAM
    -0.61
     كومونز
    -0.61
     AssemblyTitle
    -0.60
    POSITIVE LOGITS
    rxjs
    0.63
     hasilnya
    0.51
     Picchu
    0.48
    enumi
    0.48
     takers
    0.47
    }`).
    0.47
    %)$
    0.47
    awaiter
    0.47
    }`)
    0.46
    reactivex
    0.46
    Activations Density 0.020%

    No Known Activations