© 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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 14315
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to extreme actions and choices

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

    Sacrifice, cost, or willingness

    np_acts-logits-general · gemini-2.0-flash

    sacrifice or cost

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 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
    BeginInit
    -0.58
    :✨
    -0.55
    曖昧さ回避
    -0.54
     nahilalakip
    -0.52
     ujednoznacz
    -0.52
     wireType
    -0.51
    楽部
    -0.49
     вікісторінку
    -0.49
     queſta
    -0.49
    itename
    -0.47
    POSITIVE LOGITS
     sacrifice
    0.63
     sacrifices
    0.62
    sacrifice
    0.56
    sacrific
    0.54
     sacrificing
    0.53
     sacrific
    0.52
     Sacrific
    0.51
     Sacrifice
    0.50
     sacrificio
    0.49
     sacrificed
    0.49
    Activations Density 0.355%

    No Known Activations