© 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. 14388
    Prev
    Next
    INDEX
    Explanations

    punctuation marks, particularly quotation marks and semicolons

    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
    478
    +0.20
    1.1%
    38
    +0.10
    0.5%
    86
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    8
    +0.20
    0.03
    86
    +0.10
    0.03
    168
    +0.10
    0.03
    Negative Logits
    ĸ
    -2.67
    ²
    -2.66
    ĻĤ
    -2.57
    £
    -2.54
    ¿½
    -2.49
    Ļ
    -2.36
    ↵↵↵
    -2.35
    ↵
    -2.35
                  ↵
    -2.35
    -2.35
    POSITIVE LOGITS
     accord
    1.56
     lieu
    1.52
     git
    1.50
     yaml
    1.47
     sake
    1.46
    UX
    1.46
     see
    1.41
     abstraction
    1.39
     const
    1.38
     constru
    1.38
    Activations Density 0.158%

    No Known Activations