© 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. 32332
    Prev
    Next
    INDEX
    Explanations

    instances of the word "divided."

    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
    378
    +0.13
    0.7%
    196
    +0.12
    0.7%
    360
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    359
    +0.13
    0.02
    378
    +0.12
    0.01
    196
    +0.11
    0.01
    Negative Logits
    bellum
    -1.68
    ?’
    -1.57
    !’
    -1.54
    ”),
    -1.49
    .’”
    -1.47
    llo
    -1.43
    OLD
    -1.42
    ãģ§ãģį
    -1.42
     always
    -1.39
    ”)
    -1.38
    POSITIVE LOGITS
     ![](
    1.52
    enso
    1.48
    iazep
    1.47
    ert
    1.44
    iative
    1.43
    ferentially
    1.42
    utively
    1.33
    versed
    1.30
    umbent
    1.29
    ifferent
    1.27
    Activations Density 0.113%

    No Known Activations