© 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. 15355
    Prev
    Next
    INDEX
    Explanations

    references to video downloading and conversion software

    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
    188
    +0.16
    0.9%
    263
    +0.13
    0.7%
    135
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    188
    +0.16
    0.03
    469
    +0.13
    0.09
    263
    +0.11
    0.11
    Negative Logits
     disulfide
    -2.15
     typh
    -1.88
     beet
    -1.71
    prud
    -1.63
     Cys
    -1.62
     Diabetes
    -1.62
     conjug
    -1.59
     islets
    -1.59
     diabetes
    -1.56
     diab
    -1.56
    POSITIVE LOGITS
     video
    4.24
     Video
    4.19
     audio
    4.14
     playback
    4.13
     videos
    4.02
    video
    3.90
     YouTube
    3.79
     Audio
    3.79
    Video
    3.63
     streaming
    3.60
    Activations Density 1.780%

    No Known Activations