© 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. 9237
    Prev
    Next
    INDEX
    Explanations

    elements related to formatting and linking stylesheets in HTML

    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
    191
    +0.13
    0.7%
    312
    +0.13
    0.7%
    93
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    255
    +0.13
    0.00
    203
    +0.13
    0.03
    191
    +0.13
    0.04
    Negative Logits
    arner
    -1.74
    usiness
    -1.64
     Transaction
    -1.57
    nier
    -1.52
    achine
    -1.52
     pleth
    -1.47
    akov
    -1.45
    hole
    -1.45
    Decided
    -1.43
    rolet
    -1.42
    POSITIVE LOGITS
    jquery
    2.58
    css
    2.23
    stylesheet
    2.17
    styles
    2.01
    script
    2.00
    document
    1.95
    bootstrap
    1.95
     css
    1.93
    Style
    1.91
     jQuery
    1.90
    Activations Density 0.226%

    No Known Activations