© 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. 15134
    Prev
    Next
    INDEX
    Explanations

    legal and judicial terminology related to court proceedings

    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
    27
    +0.13
    0.8%
    78
    +0.12
    0.7%
    235
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    168
    +0.13
    0.05
    340
    +0.12
    0.02
    27
    +0.12
    0.04
    Negative Logits
    ureus
    -1.59
     simplest
    -1.57
    ises
    -1.57
    equation
    -1.54
    fered
    -1.46
    ylene
    -1.42
     favour
    -1.41
    resis
    -1.38
    gem
    -1.35
    ikh
    -1.34
    POSITIVE LOGITS
     Atty
    2.60
     Attorney
    2.44
     Attorneys
    2.42
     Counsel
    2.29
     Clerk
    2.07
     Judges
    2.02
     Director
    2.00
     attorney
    1.89
     counsel
    1.88
     Judge
    1.88
    Activations Density 0.173%

    No Known Activations