© 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. 8885
    Prev
    Next
    INDEX
    Explanations

    statements reflecting personal beliefs and opinions

    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
    307
    +0.31
    1.9%
    306
    +0.16
    0.9%
    186
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    269
    +0.31
    0.17
    186
    +0.16
    0.10
    108
    +0.13
    0.19
    Negative Logits
    <|outofrange|>
    -9.46
           ↵
    -9.46
    ↵↵
    -9.46
    č↵
    -9.46
    -9.46
    ↵
    -9.46
    <|outofrange|>
    -9.46
    ↵↵
    -9.46
    č↵
    -9.46
    -9.46
    POSITIVE LOGITS
    fecture
    1.42
     when
    1.40
    into
    1.33
     onward
    1.30
     ending
    1.30
    .);
    1.29
    alian
    1.28
    rile
    1.27
     onwards
    1.25
    .).
    1.25
    Activations Density 4.915%

    No Known Activations