© 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. 15253
    Prev
    Next
    INDEX
    Explanations

    questions and responses involving interactions, particularly in dialogue or conversational contexts

    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
    231
    +0.10
    0.6%
    400
    +0.09
    0.5%
    89
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    363
    +0.10
    0.03
    484
    +0.09
    0.01
    61
    +0.09
    0.03
    Negative Logits
    Ŀ
    -3.47
    ľĵ
    -3.33
    ¡
    -3.27
    ´
    -3.24
    ij
    -3.22
    ¢
    -3.19
    Ģ
    -3.11
    £
    -3.08
    Ħ
    -3.05
    ĨĴ
    -3.01
    POSITIVE LOGITS
     Do
    3.29
     do
    3.02
    do
    2.92
    Do
    2.60
     DO
    2.54
    DO
    2.21
     done
    2.15
    doing
    2.08
     Does
    2.03
     Did
    1.99
    Activations Density 0.789%

    No Known Activations