© 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. 32229
    Prev
    Next
    INDEX
    Explanations

    statistical and analytical terms related to research data

    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/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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    271
    +0.16
    0.9%
    97
    +0.12
    0.7%
    481
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    271
    +0.16
    -0.03
    167
    +0.12
    0.06
    66
    +0.11
    0.10
    Negative Logits
    ";
    -1.95
    .";
    -1.75
    essential
    -1.67
    .”
    -1.65
    ↵	↵
    -1.61
    ”.
    -1.57
    ';
    -1.56
    <>();
    -1.46
    ients
    -1.42
    ”;
    -1.39
    POSITIVE LOGITS
    ))^
    1.71
    !),
    1.68
    ),
    1.64
    ))/
    1.59
    ,)
    1.54
    )--
    1.50
    manship
    1.48
    ).)
    1.48
    *),
    1.48
    ĻĤ
    1.47
    Activations Density 2.589%

    No Known Activations