© 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. 9567
    Prev
    Next
    INDEX
    Explanations

    phrases related to health, wellness, and performance improvement in golf

    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
    273
    +0.15
    0.8%
    368
    +0.12
    0.7%
    404
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    404
    +0.15
    0.02
    368
    +0.12
    0.02
    112
    +0.12
    0.04
    Negative Logits
     Jedi
    -2.19
     conjugation
    -1.82
     Luke
    -1.81
    obacterium
    -1.75
     Planck
    -1.71
     BCG
    -1.70
     methane
    -1.69
     mAb
    -1.68
     Xml
    -1.64
     phage
    -1.63
    POSITIVE LOGITS
     Golf
    4.70
     golf
    4.66
     gol
    3.93
     tee
    3.13
     tennis
    3.01
     Woods
    2.90
     Tiger
    2.79
     tournaments
    2.77
     greens
    2.74
     Tournament
    2.67
    Activations Density 0.551%

    No Known Activations