© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 26236
    Prev
    Next
    INDEX
    Explanations

    instances of the word "government" and related terms

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/2-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.2.hook_resid_post
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    434
    +0.19
    1.1%
    186
    +0.16
    0.9%
    365
    +0.16
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    434
    +0.19
    0.06
    365
    +0.16
    0.04
    418
    +0.16
    0.01
    Negative Logits
     sexes
    -1.78
    aways
    -1.74
    )](#
    -1.68
    teenth
    -1.63
     adulthood
    -1.59
    heit
    -1.58
     someone
    -1.55
     lonely
    -1.55
     menstru
    -1.54
     alleles
    -1.51
    POSITIVE LOGITS
     contractor
    1.96
     agencies
    1.95
    bug
    1.88
     affairs
    1.87
     contracting
    1.85
     shutdown
    1.76
    committee
    1.74
     Affairs
    1.71
     officials
    1.69
     financing
    1.69
    Activations Density 0.228%

    No Known Activations