© 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. 27995
    Prev
    Next
    INDEX
    Explanations

    sections of text that contain ellipses or interruptions in thought

    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
    478
    +0.19
    1.1%
    234
    +0.12
    0.7%
    156
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    445
    +0.19
    0.02
    147
    +0.12
    0.04
    5
    +0.12
    0.04
    Negative Logits
    ellar
    -1.95
    ickness
    -1.75
    ault
    -1.74
    omy
    -1.70
    yst
    -1.67
    ા
    -1.66
    itures
    -1.65
    spr
    -1.65
    "}](#
    -1.64
    als
    -1.59
    POSITIVE LOGITS
     Continue
    2.15
    unnumbered
    2.12
    unreadable
    1.75
     More
    1.60
    Continue
    1.55
    more
    1.55
     MORE
    1.53
     </
    1.52
    etc
    1.51
     ears
    1.44
    Activations Density 0.158%

    No Known Activations