© 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-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 1287
    Prev
    Next
    INDEX
    Explanations

    structured timelines and significant events in historical contexts

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

    "at the"

    np_acts-logits-general · gemini-2.0-flash

    past time reference

    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
     GenerationType
    -0.61
    BASEPATH
    -0.60
     UIFont
    -0.60
    ]--;
    -0.59
    PerformLayout
    -0.58
    ък
    -0.57
    ]")]
    -0.57
    InSection
    -0.55
    iosis
    -0.54
    eph
    -0.54
    POSITIVE LOGITS
     damals
    1.92
    当时
    1.72
    當時
    1.68
    当時は
    1.62
    当时的
    1.61
    当時
    1.60
     damaligen
    1.53
     тогда
    1.48
    那时候
    1.42
    当時の
    1.41
    Activations Density 0.491%

    No Known Activations