© 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. 26386
    Prev
    Next
    INDEX
    Explanations

    phrases indicating common or typical behaviors and occurrences

    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
    155
    +0.14
    0.8%
    87
    +0.12
    0.7%
    23
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    155
    +0.14
    0.04
    476
    +0.12
    0.03
    404
    +0.11
    0.03
    Negative Logits
    Ħ
    -3.37
    Ĭ
    -3.35
    ĭ
    -3.09
    İ
    -3.04
    IJ
    -3.01
    Ĩ
    -2.87
    į
    -2.82
    ¾
    -2.79
    ±
    -2.77
    Į
    -2.72
    POSITIVE LOGITS
    itzerland
    1.90
    apest
    1.68
     prefer
    1.66
    ischer
    1.64
    ){#
    1.54
    bbe
    1.52
     avoided
    1.50
    itan
    1.48
    gium
    1.47
    ischen
    1.46
    Activations Density 0.093%

    No Known Activations