© 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. 14161
    Prev
    Next
    INDEX
    Explanations

    specific identifiers or keywords related to technical specifications and triggers for programming functions

    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
    161
    +0.18
    1.1%
    293
    +0.15
    0.8%
    243
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    156
    +0.18
    0.28
    111
    +0.15
    0.24
    9
    +0.14
    0.21
    Negative Logits
     sworn
    -1.73
     personal
    -1.51
     caption
    -1.42
    vant
    -1.41
    *âĢIJ
    -1.41
     concurring
    -1.37
     published
    -1.36
     Quran
    -1.32
     Constitution
    -1.28
     opinion
    -1.27
    POSITIVE LOGITS
    ľĵ
    1.66
    tone
    1.61
    ints
    1.48
    lets
    1.47
    umps
    1.46
    oker
    1.44
     cords
    1.44
    pering
    1.42
    off
    1.41
    oles
    1.38
    Activations Density 5.885%

    No Known Activations