© 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-2-2B
    7. Residual Stream - 16k
    9. 18-GEMMASCOPE-RES-16K
    11. 10687
    Prev
    Next
    INDEX
    Explanations

    actions related to food preparation and serving techniques

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

    Actions involving hands or tools

    np_acts-logits-general · gemini-2.0-flash

    gentle manipulation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     weathering
    -0.58
     fubject
    -0.56
     Bunk
    -0.56
     Baus
    -0.55
     outsourced
    -0.54
    Tembelea
    -0.54
     Kochen
    -0.53
     thrived
    -0.51
    ględ
    -0.51
    rboles
    -0.51
    POSITIVE LOGITS
     gently
    0.69
     technique
    0.67
    用力
    0.64
     Gently
    0.64
     gentle
    0.64
    WriteBarrier
    0.63
     délic
    0.60
     pulling
    0.60
     pull
    0.59
    effort
    0.58
    Activations Density 0.509%

    No Known Activations