© 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. 14376
    Prev
    Next
    INDEX
    Explanations

    technical terms and phrases related to programming and data structures

    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
    156
    +0.19
    1.1%
    237
    +0.16
    0.9%
    123
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    237
    +0.19
    0.17
    480
    +0.16
    0.10
    123
    +0.12
    0.14
    Negative Logits
     surprises
    -1.53
     miracles
    -1.43
     degrees
    -1.43
     better
    -1.36
     distinction
    -1.36
     supp
    -1.28
     SOC
    -1.23
     amaz
    -1.23
     credibility
    -1.23
     chore
    -1.22
    POSITIVE LOGITS
    merged
    1.46
    owned
    1.46
    porter
    1.44
    ho
    1.40
    ^](#
    1.39
    aque
    1.33
    osterone
    1.31
    pector
    1.31
     injured
    1.30
    hend
    1.29
    Activations Density 5.794%

    No Known Activations