© 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. 32149
    Prev
    Next
    INDEX
    Explanations

    types of opening parentheses

    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
    84
    +0.13
    0.7%
    188
    +0.10
    0.6%
    59
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    188
    +0.13
    0.02
    29
    +0.10
    0.01
    320
    +0.10
    0.02
    Negative Logits
    "'
    -1.54
    eric
    -1.33
     proportions
    -1.32
    msgstr
    -1.29
    []{
    -1.28
     included
    -1.28
    '
    -1.28
    breviations
    -1.27
     forgiven
    -1.26
    "
    -1.25
    POSITIVE LOGITS
    nier
    1.70
    ãĥ³ãĥĪ
    1.47
    ãĤ£
    1.43
    μα
    1.42
    ν
    1.40
    ÙĦ
    1.40
    ãĥ¼ãĥ«
    1.39
    nan
    1.36
    ãĥ¼
    1.35
    İ
    1.34
    Activations Density 0.043%

    No Known Activations