© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 13765
    Prev
    Next
    INDEX
    Explanations

    technical terms and logging-related concepts

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

    debugging or datetimes

    np_acts-logits-general · gemini-2.0-flash

    debug and datetime

    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
     useRef
    -0.53
    Билгалдахарш
    -0.48
    utilisons
    -0.48
    ToBounds
    -0.48
    verwijspagina
    -0.44
     ElementRef
    -0.43
    Lan
    -0.43
    billon
    -0.43
    Planes
    -0.42
     atlik
    -0.42
    POSITIVE LOGITS
    datetime
    0.72
     datetime
    0.58
    Datetime
    0.57
     surla
    0.52
     Apre
    0.52
    arangay
    0.50
     Arty
    0.50
     bề
    0.50
     argint
    0.50
    Athen
    0.50
    Activations Density 0.529%

    No Known Activations