© 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. 15280
    Prev
    Next
    INDEX
    Explanations

    references to legal or procedural terms related to evidence and court processes

    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
    83
    +0.11
    0.6%
    343
    +0.11
    0.6%
    25
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    275
    +0.11
    0.01
    359
    +0.11
    0.06
    321
    +0.11
    0.01
    Negative Logits
     theme
    -1.69
     Theme
    -1.63
     moder
    -1.56
    omorphism
    -1.55
     roles
    -1.54
     verbs
    -1.53
    unov
    -1.51
     knight
    -1.49
     virtues
    -1.49
     kings
    -1.49
    POSITIVE LOGITS
     samples
    3.74
     Samples
    3.24
     sample
    3.14
    Samples
    3.13
     specimens
    2.93
    samples
    2.89
     biopsies
    2.76
     aliqu
    2.65
     tissue
    2.64
     Sample
    2.56
    Activations Density 0.751%

    No Known Activations