© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 14064
    Prev
    Next
    INDEX
    Explanations

    repeated instances of the word "the."

    oai_token-act-pair · gpt-4o-miniTriggered 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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    21
    +0.30
    1.8%
    156
    +0.15
    0.9%
    127
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    145
    +0.30
    0.07
    291
    +0.15
    0.06
    10
    +0.13
    0.08
    Negative Logits
    º
    -2.77
    ĭ
    -2.75
    ²
    -2.73
    »
    -2.62
      ↵
    -2.61
     ↵
    -2.61
     ↵
    -2.61
    ↵            ↵
    -2.61
     ↵
    -2.61
    ↵
    -2.61
    POSITIVE LOGITS
     day
    2.37
     century
    2.13
     Month
    2.07
     world
    1.96
     same
    1.94
     month
    1.91
    same
    1.87
     week
    1.85
     latter
    1.77
     era
    1.77
    Activations Density 0.551%

    No Known Activations