© 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. 32496
    Prev
    Next
    INDEX
    Explanations

    phrases indicating a variety of opinions or experiences

    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
    490
    +0.12
    0.7%
    457
    +0.12
    0.7%
    496
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    457
    +0.12
    0.06
    118
    +0.12
    0.07
    409
    +0.11
    0.06
    Negative Logits
    sible
    -1.62
    untime
    -1.53
    ionate
    -1.52
    irtual
    -1.50
     Einstein
    -1.49
    ÃŃvel
    -1.45
    etheus
    -1.39
     magnet
    -1.36
     cube
    -1.35
    ED
    -1.33
    POSITIVE LOGITS
    Ī
    4.07
    ī
    3.52
    ↵
    3.33
    č↵
    3.33
     ↵
    3.33
    <|outofrange|>
    3.33
    ↵
    3.33
    <|outofrange|>
    3.33
    ↵  ³³³
    3.33
    ↵
    3.33
    Activations Density 0.459%

    No Known Activations