© 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. 8728
    Prev
    Next
    INDEX
    Explanations

    structured code elements and comments in programming or markup languages

    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
    78
    +0.13
    0.8%
    21
    +0.11
    0.7%
    240
    +0.09
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    255
    +0.13
    -0.00
    78
    +0.11
    0.00
    191
    +0.09
    0.02
    Negative Logits
    ment
    -1.59
    ubicin
    -1.54
    ments
    -1.38
    thon
    -1.33
    tons
    -1.32
    merce
    -1.27
    mentation
    -1.24
    tin
    -1.22
     limitation
    -1.21
    mouth
    -1.17
    POSITIVE LOGITS
    **](#
    1.46
    ĥ½
    1.39
    quo
    1.31
    |^
    1.30
    úblic
    1.25
    idenote
    1.25
    IJ
    1.25
    ________
    1.25
    }|^
    1.24
    ]^.
    1.24
    Activations Density 0.061%

    No Known Activations