© 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. 9910
    Prev
    Next
    INDEX
    Explanations

    references to claims and examples in patent literature

    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
    340
    +0.10
    0.6%
    3
    +0.10
    0.6%
    266
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    216
    +0.10
    0.04
    486
    +0.10
    0.03
    359
    +0.10
    0.03
    Negative Logits
    ](
    -1.80
    ?**
    -1.67
    istical
    -1.65
    ferroni
    -1.56
    !_
    -1.53
    cue
    -1.50
    ist
    -1.49
    ??
    -1.47
     across
    -1.46
    ?!
    -1.46
    POSITIVE LOGITS
     FIGS
    3.30
     FIG
    3.21
     Patent
    2.71
     disclosed
    2.43
     embodiments
    2.30
     WO
    2.24
     JP
    2.16
     embodiment
    2.14
     xe
    2.12
     hereinafter
    2.09
    Activations Density 0.159%

    No Known Activations