© 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. 23785
    Prev
    Next
    INDEX
    Explanations

    words that indicate ease of use or accessibility

    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
    404
    +0.12
    0.7%
    376
    +0.12
    0.6%
    146
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    114
    +0.12
    0.03
    404
    +0.12
    0.03
    146
    +0.11
    0.02
    Negative Logits
    Ķ
    -1.65
    matter
    -1.53
    Ĭ
    -1.52
    smallcaps
    -1.47
    s
    -1.46
    cember
    -1.46
    lessness
    -1.44
     form
    -1.43
    hene
    -1.42
     forms
    -1.41
    POSITIVE LOGITS
     identifiable
    1.78
     enough
    1.66
    comings
    1.56
     coincidence
    1.51
     accessing
    1.51
     accessed
    1.50
     glimps
    1.45
     seen
    1.45
     beaten
    1.45
     ?>
    1.39
    Activations Density 0.056%

    No Known Activations