© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 11127
    Prev
    Next
    INDEX
    Explanations

    mentions of diseases, especially Alzheimer's and Parkinson's, along with references to caregiving and health care issues

    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
    153
    +0.14
    0.8%
    18
    +0.12
    0.7%
    13
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    18
    +0.14
    0.09
    391
    +0.12
    0.07
    13
    +0.12
    0.07
    Negative Logits
     respective
    -1.70
    collect
    -1.41
    coding
    -1.38
    Figure
    -1.30
     dated
    -1.29
    rew
    -1.28
     youtube
    -1.26
    ogether
    -1.25
     differences
    -1.25
    boards
    -1.25
    POSITIVE LOGITS
    ics
    1.69
     attacks
    1.64
    iska
    1.60
     infection
    1.48
    burgh
    1.45
    neys
    1.42
     lightly
    1.41
    ãĤĩ
    1.40
    ney
    1.37
     himself
    1.37
    Activations Density 1.077%

    No Known Activations