© 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. 1-RES-SM
    11. 24039
    Prev
    Next
    INDEX
    Explanations

    legal citations and structured references in text

    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/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
    Neuron Alignment
    Index
    Value
    % of L₁
    478
    +0.19
    1.1%
    188
    +0.15
    0.9%
    425
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    188
    +0.19
    0.06
    425
    +0.15
    0.05
    292
    +0.13
    0.04
    Negative Logits
    opan
    -1.73
    ivity
    -1.72
     (âĪĴ
    -1.65
     supernatants
    -1.47
    os
    -1.43
     accumulated
    -1.42
    ientos
    -1.41
    oses
    -1.38
    eered
    -1.38
     notified
    -1.35
    POSITIVE LOGITS
    hereinafter
    1.95
    «
    1.87
    =>
    1.75
    .),
    1.64
     ),
    1.58
    Īĺ
    1.56
    unes
    1.56
    ĻĤ
    1.54
     =>
    1.52
     syll
    1.50
    Activations Density 0.198%

    No Known Activations