© 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. 29996
    Prev
    Next
    INDEX
    Explanations

    dates, particularly those from the 20th century

    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
    40
    +0.14
    0.8%
    292
    +0.13
    0.7%
    154
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    40
    +0.14
    0.03
    137
    +0.13
    0.03
    292
    +0.12
    0.03
    Negative Logits
     QR
    -1.56
    naments
    -1.55
     Required
    -1.52
     Control
    -1.46
    igned
    -1.45
    ções
    -1.43
     Purpose
    -1.43
    isters
    -1.42
    pecific
    -1.41
    hentication
    -1.40
    POSITIVE LOGITS
    ī
    2.47
    ĨĴ
    2.38
                  ↵
    2.27
    2.27
    č↵
    2.27
    2.27
    ↵
    2.27
       ↵
    2.27
    ↵
    2.27
    ↵
    2.27
    Activations Density 0.165%

    No Known Activations