© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 9783
    Prev
    Next
    INDEX
    Explanations

    programming syntax and structure in code

    oai_token-act-pair · gpt-4o-miniTriggered 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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    448
    +0.18
    1.0%
    203
    +0.12
    0.7%
    80
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    17
    +0.18
    0.03
    203
    +0.12
    0.03
    448
    +0.12
    0.05
    Negative Logits
    iret
    -1.87
    aucoma
    -1.75
    illery
    -1.71
     Hilton
    -1.70
    abine
    -1.65
     Kap
    -1.64
     Gren
    -1.64
     Lé
    -1.62
    Hamilton
    -1.62
     Lynch
    -1.61
    POSITIVE LOGITS
     XML
    4.26
     xml
    4.23
     Xml
    3.99
    xml
    3.52
    XML
    3.41
    Xml
    3.28
    xsl
    2.85
     DOM
    2.69
     xmlns
    2.65
    xsd
    2.50
    Activations Density 0.466%

    No Known Activations