© 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. 30120
    Prev
    Next
    INDEX
    Explanations

    error messages and syntactical issues in code or system logs

    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
    100
    +0.15
    0.8%
    365
    +0.14
    0.8%
    206
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    167
    +0.15
    0.02
    443
    +0.14
    0.03
    433
    +0.13
    0.02
    Negative Logits
    estly
    -1.61
    ulf
    -1.58
    ydr
    -1.53
    urches
    -1.51
    dale
    -1.47
    oya
    -1.45
    dern
    -1.43
     up
    -1.42
    aside
    -1.40
    adal
    -1.38
    POSITIVE LOGITS
    seek
    1.52
     description
    1.48
     wurde
    1.44
     Alert
    1.43
    gers
    1.39
     hatte
    1.37
    ¡
    1.33
    ¨
    1.33
    £
    1.32
    «
    1.31
    Activations Density 0.063%

    No Known Activations