© 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. 30140
    Prev
    Next
    INDEX
    Explanations

    references to regulations, details, and various issues related to market conditions and compliance

    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
    23
    +0.15
    0.8%
    369
    +0.15
    0.8%
    376
    +0.14
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    166
    +0.15
    0.07
    184
    +0.15
    0.09
    287
    +0.14
    0.09
    Negative Logits
    ypse
    -1.76
    etc
    -1.64
    izabeth
    -1.63
    kets
    -1.57
     etc
    -1.51
    doms
    -1.49
    kubernetes
    -1.48
    ptions
    -1.47
    verter
    -1.46
    alls
    -1.45
    POSITIVE LOGITS
     caution
    1.63
     reluctance
    1.45
     Comments
    1.42
     deference
    1.38
     tight
    1.34
     strained
    1.33
     reass
    1.31
     pronounced
    1.29
     epit
    1.27
     steep
    1.26
    Activations Density 1.451%

    No Known Activations