© 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. 24045
    Prev
    Next
    INDEX
    Explanations

    image file types, specifically PNG and PDF formats

    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
    269
    +0.16
    0.9%
    230
    +0.14
    0.8%
    46
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    230
    +0.16
    0.02
    269
    +0.14
    0.02
    490
    +0.11
    0.01
    Negative Logits
     consideration
    -1.77
    ologically
    -1.59
     due
    -1.56
    yon
    -1.55
     controlling
    -1.53
     suppose
    -1.51
     observed
    -1.49
     exploration
    -1.45
     reorganization
    -1.42
     principal
    -1.40
    POSITIVE LOGITS
    ERTY
    1.82
    INGS
    1.82
    çī
    1.71
    illion
    1.60
    icture
    1.55
    lite
    1.54
    ixel
    1.53
    flash
    1.51
    skirts
    1.50
    sburg
    1.48
    Activations Density 0.102%

    No Known Activations