© 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. 29927
    Prev
    Next
    INDEX
    Explanations

    references to a male subject or protagonist

    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
    54
    +0.13
    0.7%
    38
    +0.12
    0.6%
    447
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    89
    +0.13
    0.01
    284
    +0.12
    0.01
    500
    +0.11
    0.01
    Negative Logits
    ked
    -1.61
    mos
    -1.50
     products
    -1.43
    ]{},
    -1.38
     films
    -1.33
     pharmaceutical
    -1.33
    *âĢIJ
    -1.33
    –
    -1.25
     Tyson
    -1.24
    iro
    -1.24
    POSITIVE LOGITS
    ivals
    1.75
    odes
    1.73
    ifier
    1.71
    oret
    1.52
    rast
    1.52
    oslav
    1.51
    arily
    1.50
    :%
    1.48
     '%
    1.46
    ilities
    1.43
    Activations Density 0.001%

    No Known Activations