© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 32363
    Prev
    Next
    INDEX
    Explanations

    structured programming elements related to classes and methods in Java code

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    322
    +0.14
    0.8%
    337
    +0.11
    0.6%
    75
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    80
    +0.14
    0.09
    322
    +0.11
    0.03
    209
    +0.11
    0.09
    Negative Logits
    )\]
    -1.74
    )\].
    -1.73
    ctomy
    -1.72
     charts
    -1.57
     ([@
    -1.51
    .**
    -1.50
    ale
    -1.46
     visualization
    -1.39
    .\
    -1.36
    _(
    -1.33
    POSITIVE LOGITS
    Īĺ
    2.15
    Ĥ¬
    2.09
    ģ
    2.06
    ±
    2.02
    º
    1.88
    ĥ½
    1.86
     onCreate
    1.85
     java
    1.79
    ¢
    1.75
     ArrayList
    1.75
    Activations Density 1.004%

    No Known Activations