© 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. 2-RES-SM
    11. 9650
    Prev
    Next
    INDEX
    Explanations

    repeated instances of the word "ar."

    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/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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    450
    +0.15
    0.9%
    156
    +0.14
    0.8%
    350
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    350
    +0.15
    0.02
    450
    +0.14
    0.02
    97
    +0.13
    0.02
    Negative Logits
     respective
    -1.65
    ers
    -1.64
    ictionary
    -1.61
    ETHOD
    -1.59
    xture
    -1.56
    rapeut
    -1.54
    elist
    -1.48
    CDATA
    -1.45
    erm
    -1.44
    erv
    -1.40
    POSITIVE LOGITS
    ry
    1.75
    men
    1.70
    ctic
    1.66
    manship
    1.60
    rived
    1.56
    thur
    1.55
    ships
    1.51
    nett
    1.49
    iso
    1.45
    doms
    1.45
    Activations Density 0.083%

    No Known Activations