© 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. 30074
    Prev
    Next
    INDEX
    Explanations

    occurrences of property definitions in code

    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
    467
    +0.14
    0.8%
    275
    +0.12
    0.7%
    41
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    467
    +0.14
    0.02
    379
    +0.12
    0.02
    506
    +0.11
    0.02
    Negative Logits
    ½
    -1.95
    ¹
    -1.79
    ŀ
    -1.65
    ĨĴ
    -1.60
    IJ
    -1.58
    Ĵ
    -1.56
    Ł
    -1.55
    į
    -1.50
    cause
    -1.45
    ↵
    -1.45
    POSITIVE LOGITS
    ctin
    1.65
    ths
    1.56
    lated
    1.52
     details
    1.52
    ters
    1.50
    lections
    1.49
    umns
    1.46
    ayout
    1.45
    isters
    1.45
    olves
    1.42
    Activations Density 0.112%

    No Known Activations