© 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. 1-RES-SM
    11. 23798
    Prev
    Next
    INDEX
    Explanations

    sentences that have punctuation marks or specific sentence-ending words

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    202
    +0.12
    0.6%
    254
    +0.11
    0.6%
    185
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.12
    0.04
    156
    +0.11
    0.02
    271
    +0.10
    0.02
    Negative Logits
    nd
    -1.67
    onson
    -1.60
    amp
    -1.58
    ers
    -1.55
     than
    -1.51
    amping
    -1.49
    ksen
    -1.42
    auf
    -1.42
    behind
    -1.40
    dered
    -1.40
    POSITIVE LOGITS
    manship
    1.55
    lish
    1.43
    ypse
    1.38
     acquired
    1.38
     validity
    1.37
     chromosome
    1.35
    www
    1.32
     boundaries
    1.28
    aments
    1.28
     inserted
    1.27
    Activations Density 0.009%

    No Known Activations