© 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. 30030
    Prev
    Next
    INDEX
    Explanations

    abbreviations, case-sensitive letters, and various formatting elements related to legal texts

    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
    369
    +0.18
    1.0%
    410
    +0.17
    0.9%
    373
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    369
    +0.18
    0.12
    410
    +0.17
    0.10
    373
    +0.12
    0.09
    Negative Logits
     {}
    -1.98
     oneself
    -1.75
    lass
    -1.57
    ...]
    -1.54
     etc
    -1.54
    king
    -1.50
     UK
    -1.50
    roc
    -1.47
     centre
    -1.46
     Britain
    -1.40
    POSITIVE LOGITS
    ĥ½
    1.86
     liable
    1.79
    edly
    1.77
     filed
    1.76
     testified
    1.70
    imony
    1.69
     ÂĹ
    1.69
    Ģ
    1.64
     tresp
    1.63
    ÂĹ
    1.61
    Activations Density 0.734%

    No Known Activations