© 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. 14330
    Prev
    Next
    INDEX
    Explanations

    phrases or sections that involve comparisons and evaluations related to terms or categories

    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/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
    Neuron Alignment
    Index
    Value
    % of L₁
    471
    +0.14
    0.8%
    355
    +0.10
    0.6%
    24
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    471
    +0.14
    0.02
    55
    +0.10
    0.02
    249
    +0.10
    0.02
    Negative Logits
    atre
    -1.52
    blogger
    -1.51
    plants
    -1.43
    stituted
    -1.43
    rophe
    -1.42
     heartbeat
    -1.41
     petition
    -1.37
     affidavit
    -1.35
     trailer
    -1.34
    vdots
    -1.32
    POSITIVE LOGITS
    pace
    1.64
    ymmetric
    1.58
    lement
    1.56
     of
    1.53
    paces
    1.52
    ī
    1.50
    ão
    1.46
    uo
    1.46
    ström
    1.42
    well
    1.41
    Activations Density 0.117%

    No Known Activations