© 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. 10421
    Prev
    Next
    INDEX
    Explanations

    patterns or structures in legal contexts

    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
    21
    +0.19
    1.1%
    62
    +0.11
    0.6%
    48
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    219
    +0.19
    0.04
    200
    +0.11
    0.03
    62
    +0.11
    0.02
    Negative Logits
    aters
    -1.51
    htra
    -1.48
    soever
    -1.47
     Dyn
    -1.37
    ajas
    -1.34
     office
    -1.33
     standards
    -1.33
    enda
    -1.31
     Globe
    -1.30
    etimes
    -1.29
    POSITIVE LOGITS
    ¯
    2.62
    ¦
    2.50
    ·
    2.40
    IJ
    2.38
    İ
    2.36
    ´
    2.28
    į
    2.25
    ŀ
    2.21
    ...)
    2.19
    Ī
    2.17
    Activations Density 0.122%

    No Known Activations