© 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. 13431
    Prev
    Next
    INDEX
    Explanations

    licensing terms and identifiers in software documentation

    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
    161
    +0.19
    1.1%
    12
    +0.16
    1.0%
    364
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    364
    +0.19
    0.13
    330
    +0.16
    0.12
    161
    +0.15
    0.06
    Negative Logits
    while
    -1.66
    asmod
    -1.57
    """
    -1.51
    CDATA
    -1.44
    istar
    -1.41
    atement
    -1.41
    uppose
    -1.40
    inese
    -1.40
    arean
    -1.37
    iratory
    -1.36
    POSITIVE LOGITS
    kered
    1.50
     precip
    1.44
     :
    1.44
     rides
    1.41
    etime
    1.40
    ols
    1.37
    ¹
    1.35
     Bankruptcy
    1.31
    hood
    1.30
     â
    1.28
    Activations Density 4.563%

    No Known Activations