© 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
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 10483
    Prev
    Next
    INDEX
    Explanations

    discrepancies between perceived truth and reality

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

    phrases that contrast surface appearances with the hidden or actual reality, introducing corrections or revelations.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    the phrase "in reality" followed by a contrasting or clarifying statement.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    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
     pasal
    -0.42
    olera
    -0.40
    thritis
    -0.40
     mémor
    -0.40
    terase
    -0.40
     stub
    -0.39
     goû
    -0.39
     perfected
    -0.39
     COUR
    -0.39
     novels
    -0.39
    POSITIVE LOGITS
    reality
    0.70
     sebenarnya
    0.68
     egentligen
    0.66
     réalité
    0.65
     realidad
    0.64
    本当は
    0.64
     underlying
    0.62
     reality
    0.61
    Reality
    0.60
    underlying
    0.59
    Activations Density 0.247%

    No Known Activations