© 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. 24061
    Prev
    Next
    INDEX
    Explanations

    references to software updates or fixes

    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
    156
    +0.21
    1.2%
    173
    +0.12
    0.7%
    443
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    443
    +0.21
    0.01
    191
    +0.12
    0.01
    173
    +0.11
    0.01
    Negative Logits
    ĸ´
    -2.76
    Ĥ¬
    -2.47
    ¯
    -2.38
    º
    -2.31
    ĥ½
    -2.28
    ľĵ
    -2.27
    §
    -2.25
    ´
    -2.23
    Īĺ
    -2.21
    ĭ
    -2.19
    POSITIVE LOGITS
    ulations
    2.00
    maker
    1.87
    work
    1.86
    num
    1.83
    notes
    1.82
    aline
    1.74
    ifies
    1.74
    points
    1.73
    nik
    1.73
    ies
    1.72
    Activations Density 0.093%

    No Known Activations