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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 817
    Prev
    Next
    INDEX
    Explanations

    instances of deceit or false testimony

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

    Statements about lying or falsehood

    np_acts-logits-general · gemini-2.0-flash

    lying and falsehoods

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 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
    autoreleasepool
    -0.79
     AssemblyTitle
    -0.75
     onOptions
    -0.70
     AssemblyProduct
    -0.67
    })));
    -0.65
     calendriers
    -0.65
     mouseY
    -0.62
     виправивши
    -0.62
    unknownFields
    -0.61
     Spéc
    -0.60
    POSITIVE LOGITS
     false
    1.90
     lie
    1.74
     lies
    1.69
     lied
    1.69
     falsehood
    1.68
    false
    1.63
     False
    1.60
     fake
    1.59
     mentira
    1.54
    False
    1.52
    Activations Density 0.763%

    No Known Activations