© 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 EvalsExportsAPI 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. 10856
    Prev
    Next
    INDEX
    Explanations

    instances of code-related operations, particularly focusing on SQL statements and data manipulation functions

    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
    153
    +0.16
    0.9%
    188
    +0.14
    0.8%
    307
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    188
    +0.16
    0.01
    397
    +0.14
    0.04
    465
    +0.12
    0.02
    Negative Logits
    ī
    -2.52
    º
    -2.41
    ľ
    -2.40
    ª
    -2.27
    Ŀ
    -2.23
    Ĥ¬
    -2.22
    ĸ´
    -2.21
    ĨĴ
    -2.20
    ¯
    -2.16
    ĥ½
    -2.14
    POSITIVE LOGITS
     count
    3.78
     counting
    3.67
     COUNT
    3.35
     counts
    3.32
     tally
    3.29
     counted
    3.22
    count
    3.08
     Count
    2.95
    Count
    2.85
     Percentage
    2.72
    Activations Density 0.899%

    No Known Activations