© 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. 26313
    Prev
    Next
    INDEX
    Explanations

    equations or mathematical expressions involving equality

    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
    478
    +0.13
    0.7%
    147
    +0.12
    0.7%
    358
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    147
    +0.13
    0.02
    266
    +0.12
    0.01
    191
    +0.12
    0.01
    Negative Logits
    idine
    -1.74
     yourself
    -1.66
     manually
    -1.62
     onto
    -1.59
     herself
    -1.58
    ously
    -1.56
     down
    -1.52
    isin
    -1.50
     myself
    -1.47
     differently
    -1.46
    POSITIVE LOGITS
    srep
    1.70
    aturday
    1.46
    doibase
    1.45
    opyright
    1.39
     że
    1.30
    orns
    1.26
    opsis
    1.25
    aggio
    1.24
    amais
    1.22
    reptococcus
    1.22
    Activations Density 0.034%

    No Known Activations