© 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. 3990
    Prev
    Next
    INDEX
    Explanations

    terms related to technical specifications and performance metrics

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

    Units of measure or quantities

    np_acts-logits-general · gemini-2.0-flash

    units measurements and finance

    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
     snippetHide
    -0.58
     nonatomic
    -0.55
    makeText
    -0.55
     surla
    -0.54
    __':
    -0.53
    postsleuth
    -0.53
    endphp
    -0.52
    URLException
    -0.52
    __":
    -0.50
    RTCF
    -0.49
    POSITIVE LOGITS
     pleaſure
    0.49
     ſta
    0.43
     indígen
    0.41
     diſt
    0.40
     escol
    0.40
    enregist
    0.40
     âgées
    0.39
     deſt
    0.39
    spoiler
    0.38
    StructEnd
    0.38
    Activations Density 0.683%

    No Known Activations