© 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. 2-RES-SM
    11. 14343
    Prev
    Next
    INDEX
    Explanations

    technical specifications and parameters related to software and programming

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    320
    +0.17
    1.0%
    442
    +0.15
    0.9%
    17
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    153
    +0.17
    0.07
    320
    +0.15
    0.13
    342
    +0.15
    0.06
    Negative Logits
    ,”
    -1.83
     [@
    -1.67
    ():
    -1.67
     together
    -1.66
    ”?
    -1.65
    ().
    -1.59
    .^\[[@
    -1.59
    ?”
    -1.53
    .[@
    -1.52
    [@
    -1.51
    POSITIVE LOGITS
     sails
    1.67
    rients
    1.34
    aras
    1.27
    ylvania
    1.25
     tiempo
    1.23
     defect
    1.21
    ama
    1.21
    akes
    1.20
    itars
    1.20
    oxin
    1.20
    Activations Density 5.858%

    No Known Activations