© 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. 2-RES-SM
    11. 27086
    Prev
    Next
    INDEX
    Explanations

    references to sports and competition

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    156
    +0.30
    1.8%
    186
    +0.21
    1.2%
    263
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    186
    +0.30
    0.07
    137
    +0.21
    0.13
    261
    +0.15
    0.03
    Negative Logits
    arily
    -1.45
    ographically
    -1.44
    aneously
    -1.43
     facie
    -1.39
    ochemical
    -1.36
     thanks
    -1.35
     Circuit
    -1.32
    ounces
    -1.31
    cember
    -1.31
    ÑĪе
    -1.30
    POSITIVE LOGITS
    ĻĤ
    2.92
    ĭ
    2.62
    ĨĴ
    2.50
    į
    2.15
    Ħ
    2.09
    ĸ
    2.07
    ī
    2.07
    ĺ
    2.06
    Ķ
    2.00
    Ĥ
    1.98
    Activations Density 2.389%

    No Known Activations