© 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. 24-GEMMASCOPE-RES-16K
    6. 2662
    Prev
    Next
    INDEX
    Explanations

    phrases related to scientific analyses and experimental procedures

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

    Follows "were" or "was" and precedes a verb

    np_acts-logits-general · gemini-2.0-flash

    past tense performed actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    PhysRevLett
    -0.55
    MemoryWarning
    -0.54
    IBarButtonItem
    -0.53
    gham
    -0.53
    öffentlichungen
    -0.51
    hasia
    -0.50
    avice
    -0.50
    enumi
    -0.49
    orama
    -0.49
    setContentView
    -0.49
    POSITIVE LOGITS
     performed
    2.40
     Performed
    2.01
     conducted
    1.94
    performed
    1.90
     perform
    1.73
    Performed
    1.67
     PERFORM
    1.65
    Perform
    1.65
     Perform
    1.63
     executed
    1.63
    Activations Density 0.833%

    No Known Activations