© 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. 4-RES-SM
    11. 15124
    Prev
    Next
    INDEX
    Explanations

    phrases related to surgical procedures and medical conditions

    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/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
    Neuron Alignment
    Index
    Value
    % of L₁
    9
    +0.26
    1.6%
    397
    +0.16
    1.0%
    78
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    9
    +0.26
    0.10
    111
    +0.16
    0.12
    156
    +0.15
    0.14
    Negative Logits
    .*]{},
    -1.80
    .^[@
    -1.74
    .*;
    -1.72
    .).
    -1.70
    ).
    -1.65
    }?
    -1.65
    ).[
    -1.65
    .);
    -1.61
    ){#
    -1.61
    ,\\
    -1.60
    POSITIVE LOGITS
           ↵
    5.36
    <|outofrange|>
    5.36
    5.36
    <|outofrange|>
    5.36
    ↵
    5.36
    5.36
    5.36
    5.36
    5.36
    č↵č↵
    5.36
    Activations Density 2.379%

    No Known Activations