© 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. 4-RES-SM
    11. 8801
    Prev
    Next
    INDEX
    Explanations

    aspects related to craftsmanship and quality materials

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    307
    +0.21
    1.2%
    306
    +0.19
    1.1%
    466
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    307
    +0.21
    0.10
    306
    +0.19
    0.06
    466
    +0.14
    0.09
    Negative Logits
    ĨĴ
    -2.51
    »
    -2.39
    Ł
    -2.31
    ¢
    -2.29
    ħ
    -2.28
    ī
    -2.27
    ¨
    -2.24
    ŀ
    -2.21
    Ń
    -2.20
    ı
    -2.16
    POSITIVE LOGITS
     recycled
    1.76
     traditional
    1.50
     patented
    1.48
     steel
    1.47
     proprietary
    1.46
     cardboard
    1.45
     talented
    1.44
     polyethylene
    1.43
     engineered
    1.42
     stainless
    1.40
    Activations Density 1.156%

    No Known Activations