© 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. 1-RES-SM
    11. 24421
    Prev
    Next
    INDEX
    Explanations

    instances of the word "clean" and its variants

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    198
    +0.15
    0.9%
    173
    +0.15
    0.9%
    424
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    198
    +0.15
    0.03
    173
    +0.15
    0.03
    424
    +0.12
    0.03
    Negative Logits
    hog
    -1.47
    osp
    -1.46
    romycin
    -1.42
    ÑĤо
    -1.42
    resis
    -1.40
    uses
    -1.38
     subsequ
    -1.35
    ÑĤоÑĢ
    -1.34
    clerosis
    -1.34
    gian
    -1.29
    POSITIVE LOGITS
    liness
    2.09
    ening
    1.91
    ener
    1.84
    eners
    1.75
    cloth
    1.59
    shop
    1.59
    ups
    1.57
    ies
    1.55
     appliances
    1.54
    erals
    1.50
    Activations Density 0.127%

    No Known Activations