© 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. 23852
    Prev
    Next
    INDEX
    Explanations

    numerical codes or identifiers

    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
    246
    +0.15
    0.8%
    502
    +0.13
    0.7%
    301
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    116
    +0.15
    0.03
    408
    +0.13
    0.03
    151
    +0.13
    0.03
    Negative Logits
     closer
    -1.68
     reversal
    -1.65
     assumptions
    -1.53
    aken
    -1.51
    comings
    -1.49
    ually
    -1.49
     realities
    -1.48
    ivity
    -1.47
     assumption
    -1.46
     logar
    -1.44
    POSITIVE LOGITS
    ĸ
    3.08
    ¦
    2.80
    ¢
    2.70
    ¼
    2.66
    IJ
    2.62
    ¹
    2.47
    ¶
    2.46
    Ĵ
    2.45
    Ķ
    2.41
    ³
    2.36
    Activations Density 0.084%

    No Known Activations