© 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. 8822
    Prev
    Next
    INDEX
    Explanations

    strings or parts of words that share similarities or patterns

    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/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
    Neuron Alignment
    Index
    Value
    % of L₁
    257
    +0.13
    0.7%
    326
    +0.12
    0.7%
    141
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    257
    +0.13
    0.03
    148
    +0.12
    0.02
    394
    +0.12
    0.02
    Negative Logits
    emente
    -1.76
     NOTICE
    -1.76
    mente
    -1.73
    ery
    -1.70
    edly
    -1.59
    ja
    -1.59
    eer
    -1.57
     ever
    -1.56
    gmail
    -1.53
    iei
    -1.51
    POSITIVE LOGITS
    ategy
    3.07
    ateg
    2.56
    ¦
    2.52
    ĺ
    2.41
    ¡
    2.36
    Ĵ
    2.34
    ½
    2.31
    Ħ
    2.27
    Ļª
    2.27
    Ļ
    2.10
    Activations Density 0.109%

    No Known Activations