© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 11723
    Prev
    Next
    INDEX
    Explanations

    descriptive phrases that denote actions and interactions in narratives

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

    Followed by a verb of action/manipulation

    np_acts-logits-general · gemini-2.0-flash

    grabbing, pulling, or throwing

    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
     fédé
    -0.56
     pungkas
    -0.56
     domés
    -0.56
     térm
    -0.53
     hosting
    -0.52
    sells
    -0.51
    saraba
    -0.50
     périph
    -0.49
    sä
    -0.47
     animé
    -0.47
    POSITIVE LOGITS
     flick
    0.86
     ostavi
    0.78
     stroking
    0.77
     put
    0.72
     strum
    0.71
    用手
    0.71
     grabbing
    0.69
     pull
    0.69
     pulling
    0.69
     swat
    0.69
    Activations Density 0.572%

    No Known Activations