© 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. 2-RES-SM
    11. 32272
    Prev
    Next
    INDEX
    Explanations

    parentheses in the 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/2-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.2.hook_resid_post
    Hook Layer
    2
    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%
    23
    +0.16
    0.9%
    489
    +0.15
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    188
    +0.19
    0.06
    219
    +0.16
    0.06
    489
    +0.15
    0.03
    Negative Logits
     awards
    -1.79
     (“
    -1.70
     scholars
    -1.64
    issance
    -1.64
    itud
    -1.63
     honors
    -1.62
     warrants
    -1.57
    etus
    -1.52
     remains
    -1.51
    ishers
    -1.48
    POSITIVE LOGITS
    ...)
    2.63
    ?)
    2.49
    >=
    2.42
    ?).
    2.06
     ),
    1.94
     ).
    1.92
    possibly
    1.90
    disambiguation
    1.78
    >\
    1.74
    ?'
    1.73
    Activations Density 0.165%

    No Known Activations