© 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. 24056
    Prev
    Next
    INDEX
    Explanations

    instructional language related to software functions and commands

    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
    96
    +0.12
    0.7%
    330
    +0.11
    0.6%
    21
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    420
    +0.12
    0.03
    89
    +0.11
    0.04
    339
    +0.10
    0.04
    Negative Logits
     ethic
    -1.65
    ime
    -1.63
    ŀ
    -1.63
     opportun
    -1.59
     era
    -1.52
    ocene
    -1.49
     estimates
    -1.49
     Ages
    -1.47
    ĥ
    -1.46
    ĵ
    -1.45
    POSITIVE LOGITS
     scrolling
    2.49
     click
    2.43
    click
    2.34
     scroll
    2.30
    Click
    2.27
     button
    2.21
     clicking
    2.21
     cursor
    2.21
     manually
    2.15
     clicked
    2.15
    Activations Density 1.508%

    No Known Activations