© 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. 29863
    Prev
    Next
    INDEX
    Explanations

    instances of the word "this" in various forms, indicating a focus on immediacy or relevance

    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
    111
    +0.12
    0.6%
    99
    +0.11
    0.6%
    11
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    402
    +0.12
    0.04
    99
    +0.11
    0.05
    39
    +0.11
    0.05
    Negative Logits
     rest
    -1.66
     tested
    -1.54
    raints
    -1.53
    oretic
    -1.50
     chains
    -1.44
     corresponding
    -1.44
    onas
    -1.43
    azard
    -1.42
    oret
    -1.42
    atre
    -1.40
    POSITIVE LOGITS
    »¿
    1.83
    Ļ
    1.58
    laws
    1.54
    «
    1.52
    ĵ
    1.45
    dating
    1.42
    ³
    1.40
    law
    1.38
    į
    1.37
    film
    1.36
    Activations Density 0.209%

    No Known Activations