© 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. 8631
    Prev
    Next
    INDEX
    Explanations

    terms and phrases related to comparisons and contrasts in 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
    172
    +0.12
    0.6%
    433
    +0.11
    0.6%
    175
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    497
    +0.12
    0.01
    88
    +0.11
    0.01
    203
    +0.11
    0.01
    Negative Logits
     HSP
    -1.71
    HO
    -1.70
     h
    -1.69
     Mississippi
    -1.66
     H
    -1.64
    H
    -1.62
     HS
    -1.62
     racial
    -1.60
    OH
    -1.59
    HS
    -1.58
    POSITIVE LOGITS
     Del
    3.89
    Del
    3.25
     DEL
    2.95
    del
    2.92
     del
    2.55
     delir
    2.30
    DEL
    2.29
     Val
    2.17
     Delaware
    2.13
     Delivery
    1.93
    Activations Density 0.990%

    No Known Activations