© 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. 14046
    Prev
    Next
    INDEX
    Explanations

    mentions of football clubs and player statistics

    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
    404
    +0.14
    0.8%
    411
    +0.12
    0.7%
    434
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    404
    +0.14
    0.07
    340
    +0.12
    0.07
    128
    +0.12
    0.06
    Negative Logits
     hemp
    -2.14
     $%
    -2.01
     fiber
    -1.98
     Plaintiffs
    -1.90
     vapor
    -1.90
     dend
    -1.90
     CDC
    -1.85
     mercury
    -1.84
     odor
    -1.84
     USDA
    -1.81
    POSITIVE LOGITS
     footballers
    3.97
     FIFA
    3.95
     Arsenal
    3.82
     Football
    3.74
     footballer
    3.69
     Liverpool
    3.63
     football
    3.62
    Football
    3.56
     midfielder
    3.53
    football
    3.43
    Activations Density 1.831%

    No Known Activations