© 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. 8637
    Prev
    Next
    INDEX
    Explanations

    processes related to order and purchase management

    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
    397
    +0.30
    1.9%
    21
    +0.16
    1.0%
    240
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    397
    +0.30
    0.10
    267
    +0.16
    0.08
    500
    +0.10
    0.06
    Negative Logits
    Īĺ
    -5.08
    ¿½
    -4.92
    Ł
    -4.88
    ŀ
    -4.87
    «
    -4.77
    »¿
    -4.74
    ®
    -4.73
    ļ
    -4.72
    Ĩ
    -4.71
    ³
    -4.67
    POSITIVE LOGITS
    ↵↵↵↵↵↵↵↵
    1.81
     {#
    1.69
    ↵↵↵
    1.66
    ↵↵↵↵↵↵↵
    1.64
    .**]{}
    1.60
    ${
    1.60
    ↵↵↵↵↵
    1.60
    ”)
    1.60
    ↵↵↵↵
    1.59
    ↵↵
    1.59
    Activations Density 2.166%

    No Known Activations