© 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. 14382
    Prev
    Next
    INDEX
    Explanations

    references to legal situations and issues regarding personal safety and well-being

    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
    410
    +0.15
    0.8%
    320
    +0.13
    0.7%
    318
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    318
    +0.15
    0.01
    470
    +0.13
    0.07
    410
    +0.11
    0.06
    Negative Logits
    ĨĴ
    -2.30
    ĻĤ
    -2.06
    Ĥ¬
    -1.75
    Ļª
    -1.66
    ){#
    -1.66
    »¿
    -1.59
    ))$.
    -1.52
    ))$
    -1.51
    )</
    -1.51
    ¯
    -1.51
    POSITIVE LOGITS
    theless
    1.89
    ruptcy
    1.71
    iative
    1.60
    edly
    1.57
    venue
    1.54
    ergic
    1.50
    recht
    1.49
    selves
    1.45
    stitution
    1.43
    isite
    1.41
    Activations Density 0.970%

    No Known Activations