© 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. 23965
    Prev
    Next
    INDEX
    Explanations

    nutritional vitamins and mineral deficiencies

    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
    103
    +0.12
    0.7%
    272
    +0.12
    0.7%
    151
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    337
    +0.12
    0.04
    417
    +0.12
    0.04
    262
    +0.11
    0.04
    Negative Logits
     unsuccessful
    -1.73
     euthan
    -1.68
     neurolog
    -1.62
     anesthetized
    -1.62
     scheduled
    -1.57
     passengers
    -1.57
     fatal
    -1.57
     deaf
    -1.54
     nausea
    -1.52
     married
    -1.50
    POSITIVE LOGITS
    aturated
    1.86
     contribution
    1.82
     footprint
    1.74
    Constants
    1.70
     synthase
    1.69
    availability
    1.62
    ylation
    1.60
    ribution
    1.59
     source
    1.54
     reductase
    1.53
    Activations Density 0.393%

    No Known Activations