© 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. 31520
    Prev
    Next
    INDEX
    Explanations

    terms related to pulmonary health and diseases

    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
    164
    +0.14
    0.8%
    267
    +0.14
    0.8%
    47
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    164
    +0.14
    0.03
    389
    +0.14
    0.03
    18
    +0.13
    0.02
    Negative Logits
     Interests
    -2.00
     interests
    -1.86
     scholarship
    -1.83
     else
    -1.75
     href
    -1.74
     websites
    -1.67
     publication
    -1.65
     friends
    -1.63
     Scholar
    -1.62
    unnumbered
    -1.61
    POSITIVE LOGITS
    monary
    1.80
    iratory
    1.79
    ICU
    1.76
     ventilation
    1.76
     smoke
    1.72
     eject
    1.65
    neum
    1.65
     air
    1.64
    ulmonary
    1.62
     airway
    1.62
    Activations Density 0.165%

    No Known Activations