© 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. 101129
    Prev
    Next
    INDEX
    Explanations

    elements related to color-changing materials and their applications

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

    color shift, shimmer, light effects

    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
    '],
    
    -0.49
    Tracce
    -0.45
    ].)
    -0.45
    ')")
    -0.44
     الرياضيه
    -0.44
    "),
    
    -0.44
    ništ
    -0.43
    Isolated
    -0.42
     **/
    
    -0.42
    "],
    
    -0.42
    POSITIVE LOGITS
    MessageTagHelper
    0.47
    AutoScaleMode
    0.43
     shimmering
    0.42
     subtle
    0.42
     contrasts
    0.40
     Савезне
    0.39
    랬
    0.39
    AppMethodBeat
    0.38
    RTEE
    0.38
    GTCX
    0.38
    Activations Density 0.019%

    No Known Activations