© 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. 24264
    Prev
    Next
    INDEX
    Explanations

    instances of the word "com" in various contexts

    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
    119
    +0.17
    0.9%
    510
    +0.10
    0.5%
    489
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    489
    +0.17
    0.03
    119
    +0.10
    0.02
    348
    +0.10
    0.03
    Negative Logits
    ureus
    -1.86
     Parenthood
    -1.63
     forgiven
    -1.46
    Indeed
    -1.40
     Redistributions
    -1.34
    CHO
    -1.33
    Continue
    -1.32
     left
    -1.31
     entire
    -1.29
    doch
    -1.29
    POSITIVE LOGITS
    ycin
    2.32
    binant
    2.08
    yll
    1.97
    pson
    1.95
    plicity
    1.78
    poser
    1.77
    neys
    1.75
    urations
    1.74
    ène
    1.74
    owitz
    1.73
    Activations Density 0.115%

    No Known Activations