© 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. 9926
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to stock and equity markets

    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
    221
    +0.14
    0.8%
    449
    +0.13
    0.7%
    496
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    262
    +0.14
    0.02
    496
    +0.13
    0.02
    133
    +0.12
    0.03
    Negative Logits
    asio
    -1.72
    iatric
    -1.64
    xture
    -1.56
    astically
    -1.56
    MODULE
    -1.55
    ropolitan
    -1.49
    icate
    -1.49
    unks
    -1.49
    static
    -1.45
    icated
    -1.44
    POSITIVE LOGITS
    pile
    3.00
    holm
    2.55
    holder
    2.54
    holders
    2.38
     exchanges
    2.32
     market
    2.21
     markets
    2.13
    yards
    2.12
     exchange
    2.09
    bro
    2.03
    Activations Density 0.297%

    No Known Activations