© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 8975
    Prev
    Next
    INDEX
    Explanations

    complex sentences discussing comparative evaluations or contrasting ideas

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

    Comes after comparative adjectives

    np_acts-logits-general · gemini-2.0-flash

    much better

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 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
    verwijspagina
    -0.70
    énario
    -0.56
     تضيفلها
    -0.55
    :✨
    -0.51
    Edited
    -0.51
     հղումներ
    -0.51
    Mutagenicity
    -0.48
    tricks
    -0.46
    AndEndTag
    -0.44
    Архівовано
    -0.44
    POSITIVE LOGITS
    onActivityResult
    0.42
    Autor
    0.40
    참고
    0.39
     bubuk
    0.37
    RunWith
    0.37
    ða
    0.35
     Autor
    0.35
     jemals
    0.35
     насеље
    0.34
     Besøkt
    0.34
    Activations Density 0.111%

    No Known Activations