© 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-2-2B
    7. Residual Stream - 16k
    9. 14-GEMMASCOPE-RES-16K
    11. 3654
    Prev
    Next
    INDEX
    Explanations

    instances of the word "lie" and its variations, indicating themes of deception or falsehood

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 14-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_14/width_16k/average_l0_84
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.14.hook_resid_post
    Hook Layer
    14
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Jacobsen
    -0.80
     McCl
    -0.78
    AndEndTag
    -0.76
    ėl
    -0.74
    рс
    -0.73
    بالإنجليزية
    -0.73
    σσ
    -0.72
    agoza
    -0.71
    urator
    -0.71
     Medford
    -0.70
    POSITIVE LOGITS
    Lie
    1.11
     lie
    1.10
     Lie
    1.02
     LIE
    0.98
    Lies
    0.96
     Lies
    0.94
     lying
    0.93
    lying
    0.89
     Lying
    0.89
     lies
    0.89
    Activations Density 0.088%

    No Known Activations