© 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. 1-RES-SM
    11. 23905
    Prev
    Next
    INDEX
    Explanations

    references to locations and events occurring there

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    15
    +0.14
    0.8%
    23
    +0.13
    0.7%
    229
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    15
    +0.14
    0.03
    229
    +0.13
    0.02
    198
    +0.13
    0.02
    Negative Logits
    ÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤ
    -4.34
     ↵
    -4.30
    -4.30
    <|outofrange|>
    -4.30
    <|outofrange|>
    -4.30
     ↵
    -4.30
    ↵
    -4.30
    čč
    -4.30
    ↵↵
    -4.30
    č↵
    -4.30
    POSITIVE LOGITS
    aterally
    1.98
    atively
    1.93
    aneously
    1.80
    ontally
    1.76
    utively
    1.73
    upon
    1.68
    vac
    1.64
    inct
    1.62
    ETHOD
    1.62
    atically
    1.61
    Activations Density 0.127%

    No Known Activations