© 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. 26295
    Prev
    Next
    INDEX
    Explanations

    references to historical events or entities related to concentration camps and societal 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
    +0.17
    1.0%
    355
    +0.14
    0.8%
    487
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    146
    +0.17
    0.05
    246
    +0.14
    0.09
    135
    +0.12
    0.07
    Negative Logits
    ↵
    -1.82
    ↵
    -1.72
    iston
    -1.53
    innings
    -1.40
    il
    -1.32
    nj
    -1.31
    acy
    -1.31
    ↵
    -1.31
     safely
    -1.28
    åºı
    -1.28
    POSITIVE LOGITS
    ī
    1.63
    ·¸
    1.51
    '?
    1.50
    Ģ
    1.47
     <-
    1.43
    ugh
    1.40
    =$(
    1.36
     =>
    1.34
    figure
    1.33
    ³
    1.31
    Activations Density 2.330%

    No Known Activations