© 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. 4-RES-SM
    11. 15158
    Prev
    Next
    INDEX
    Explanations

    themes related to political theory and the impact of governance on society

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    156
    +0.20
    1.2%
    21
    +0.18
    1.0%
    186
    +0.18
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    186
    +0.20
    0.11
    112
    +0.18
    0.09
    418
    +0.18
    0.05
    Negative Logits
    -1.93
    ↵
    -1.93
    ↵↵↵
    -1.93
    -1.93
    -1.93
                ↵
    -1.93
    <|outofrange|>
    -1.93
    -1.93
    -1.93
    -1.93
    POSITIVE LOGITS
     affairs
    2.65
    ism
    2.61
     supremacy
    2.43
    isms
    2.37
     politics
    2.36
     governance
    2.34
     organizations
    2.29
     society
    2.29
     institutions
    2.28
     disasters
    2.24
    Activations Density 1.386%

    No Known Activations