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. 2213
    Prev
    Next
    INDEX
    Explanations

    programming-related elements such as method calls and object attributes

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

    Code snippets using "this" or "obj"

    np_acts-logits-general · gemini-2.0-flash

    this in code

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     Elsewhere
    -0.55
     برانيه
    -0.51
     đàn
    -0.49
    ليق
    -0.46
     jao
    -0.45
     TestBed
    -0.44
    ดาว
    -0.43
    qtype
    -0.43
     Seem
    -0.43
    ể
    -0.43
    POSITIVE LOGITS
    ::
    0.90
    >::
    0.69
    ########.
    0.67
    ::$
    0.65
     ::
    0.64
     فريبيس
    0.64
    oa̍t
    0.60
     hảo
    0.57
    '])->
    0.57
    conscious
    0.55
    Activations Density 0.097%

    No Known Activations