© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 1-RES-SM
    11. 28291
    Prev
    Next
    INDEX
    Explanations

    importantly named entities and geographical references associated with the Middle East and Islam

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/1-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.1.hook_resid_post
    Hook Layer
    1
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    500
    +0.13
    0.7%
    410
    +0.12
    0.6%
    206
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    410
    +0.13
    0.10
    369
    +0.12
    0.05
    98
    +0.11
    -0.01
    Negative Logits
     paint
    -1.56
     coloring
    -1.54
    urable
    -1.52
     Jake
    -1.52
     Julie
    -1.49
     goog
    -1.49
     forward
    -1.47
     Pete
    -1.47
    colors
    -1.45
    ORT
    -1.45
    POSITIVE LOGITS
     Abdullah
    2.15
     Shah
    2.13
     Islamic
    2.09
     Muhammad
    2.08
     militants
    2.07
     terrorists
    2.07
     Islamist
    2.01
     Muslim
    1.95
     mosque
    1.92
     Sultan
    1.92
    Activations Density 1.161%

    No Known Activations