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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 4458
    Prev
    Next
    INDEX
    Explanations

    technical specifications and performance metrics of electronic devices

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

    Precedes "be" or "to"

    np_acts-logits-general · gemini-2.0-flash

    should be, meant to, user prompt

    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
    SharedCtor
    -0.68
    StoryboardSegue
    -0.59
     ſtate
    -0.59
    RefNanny
    -0.57
     raiſ
    -0.53
    orianCalendar
    -0.52
     ſtre
    -0.51
    endphp
    -0.51
     betweenstory
    -0.51
     cauſe
    -0.50
    POSITIVE LOGITS
     Überras
    0.42
     impressive
    0.41
     usable
    0.39
     abounds
    0.39
     underrated
    0.38
     noteworthy
    0.38
     improvements
    0.37
     Konkurrenz
    0.37
     slick
    0.37
     sleek
    0.37
    Activations Density 0.159%

    No Known Activations