© 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. 14129
    Prev
    Next
    INDEX
    Explanations

    function declarations and method signatures in programming 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
    400
    +0.17
    1.0%
    293
    +0.17
    1.0%
    302
    +0.16
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    167
    +0.17
    0.08
    377
    +0.17
    0.11
    400
    +0.16
    0.04
    Negative Logits
    pez
    -1.92
    acht
    -1.59
    ých
    -1.41
    minster
    -1.40
    chle
    -1.39
    usa
    -1.34
    ylene
    -1.33
    ollen
    -1.30
    ivat
    -1.30
    oden
    -1.29
    POSITIVE LOGITS
    IJ
    2.26
    ij
    2.22
    ¿½
    2.15
    ¼
    2.05
    ©
    2.04
    ĻĤ
    2.04
    ¤
    2.01
    ¹
    1.98
    ¥
    1.93
    ĺ
    1.90
    Activations Density 4.348%

    No Known Activations