Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 961
    Prev
    Next
    INDEX
    Explanations

    instances of the word "consensus" and related discussions about agreements or collective opinions

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

    consensus

    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
     Irán
    -0.46
     médic
    -0.38
     Turquía
    -0.36
    Woman
    -0.36
    Turkey
    -0.35
    Harry
    -0.35
     médicale
    -0.35
    Pter
    -0.35
     Turquie
    -0.35
     Egipto
    -0.35
    POSITIVE LOGITS
     Consensus
    1.36
     consensus
    1.31
    consensus
    1.18
    Consensus
    1.16
     consenso
    0.89
     consens
    0.79
    writeFieldEnd
    0.77
    ensus
    0.72
     Roskov
    0.69
     Conno
    0.66
    Activations Density 0.007%

    No Known Activations