© 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. 8676
    Prev
    Next
    INDEX
    Explanations

    comparative phrases that quantify or assess differences or similarities between groups

    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
    307
    +0.23
    1.4%
    466
    +0.16
    1.0%
    25
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    408
    +0.23
    0.04
    485
    +0.16
    0.03
    307
    +0.13
    0.05
    Negative Logits
    port
    -1.65
    github
    -1.35
    Join
    -1.30
    like
    -1.28
    iset
    -1.28
    omorphism
    -1.28
    azine
    -1.27
    isation
    -1.26
    doi
    -1.25
     offence
    -1.24
    POSITIVE LOGITS
     those
    2.62
     ours
    2.41
     others
    2.37
     other
    2.31
     theirs
    2.21
     placebo
    2.08
    ↵
    2.07
    2.07
    2.07
    <|outofrange|>
    2.07
    Activations Density 0.401%

    No Known Activations