© 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-2-9B-IT
    7. Residual Stream - 131k
    9. 31-GEMMASCOPE-RES-131K
    11. 23598
    Prev
    Next
    INDEX
    Explanations

    instances of decision-making, especially in contexts of regret or consideration

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

    pronoun or name followed by action verb

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    receive
    -0.66
     receive
    -0.66
     receives
    -0.65
     recieve
    -0.62
     Receive
    -0.59
     ricevere
    -0.58
     recie
    -0.57
    Receive
    -0.56
     recibir
    -0.56
     received
    -0.54
    POSITIVE LOGITS
     стара
    0.55
    ImageContext
    0.48
     try
    0.48
    VYMaps
    0.42
     attempt
    0.42
    あえて
    0.42
     tries
    0.42
    故意
    0.41
     choose
    0.41
     deliberately
    0.40
    Activations Density 0.117%

    No Known Activations