© 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. 30056
    Prev
    Next
    INDEX
    Explanations

    repeated sequences of the letter "k"

    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
    369
    +0.14
    0.8%
    499
    +0.11
    0.7%
    388
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    499
    +0.14
    0.04
    299
    +0.11
    0.03
    171
    +0.10
    0.04
    Negative Logits
    TRODUCTION
    -1.59
    e
    -1.56
    effective
    -1.49
    vers
    -1.38
    cess
    -1.37
    paper
    -1.36
    dtd
    -1.35
    ppings
    -1.32
    ¨
    -1.30
    cession
    -1.30
    POSITIVE LOGITS
    iewicz
    2.53
    irk
    2.31
    ruptcy
    1.83
    afka
    1.82
    owitz
    1.75
    manship
    1.74
    izumab
    1.65
    chen
    1.61
    hores
    1.61
    endas
    1.60
    Activations Density 0.106%

    No Known Activations