© 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. 30721
    Prev
    Next
    INDEX
    Explanations

    programming language syntax and structures

    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
    302
    +0.13
    0.8%
    78
    +0.13
    0.7%
    197
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    364
    +0.13
    0.09
    80
    +0.13
    0.08
    78
    +0.13
    0.04
    Negative Logits
    Ļ
    -2.50
    Ļª
    -2.46
    Ĵ
    -2.36
    Ĭ
    -2.28
    ı
    -2.18
    IJ
    -2.17
    ÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤ
    -2.16
    Īĺ
    -2.13
    ÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤ
    -2.13
    ÃĥÃĤÃĥÃĤ
    -2.08
    POSITIVE LOGITS
     occurred
    1.55
     Democr
    1.43
    ounces
    1.35
     {#
    1.29
     envelope
    1.27
     NS
    1.27
     departed
    1.26
    *.
    1.25
    ama
    1.24
    sti
    1.23
    Activations Density 3.912%

    No Known Activations