© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 72375
    Prev
    Next
    INDEX
    Explanations

    actions related to the positioning and placement of individuals or objects, especially lying down

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

    lying or reclining

    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-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    parseColor
    -0.40
     StringTokenizer
    -0.40
    TRA
    -0.40
    JST
    -0.39
     JST
    -0.38
    delimiter
    -0.38
     superstitions
    -0.38
    storms
    -0.37
     Mahat
    -0.37
    thums
    -0.36
    POSITIVE LOGITS
    躺
    0.67
     reclining
    0.67
    🛌
    0.60
    躺在
    0.59
     sleeping
    0.59
     couch
    0.59
    Beds
    0.59
     mattress
    0.59
     bed
    0.58
     mergeFrom
    0.57
    Activations Density 0.018%

    No Known Activations