© 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. 28352
    Prev
    Next
    INDEX
    Explanations

    instances of the word "two."

    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
    61
    +0.11
    0.6%
    139
    +0.11
    0.6%
    466
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    139
    +0.11
    0.05
    483
    +0.11
    0.05
    61
    +0.11
    0.05
    Negative Logits
     politics
    -1.68
    sworth
    -1.55
     activism
    -1.52
     fame
    -1.51
    äºİ
    -1.50
     wherever
    -1.50
    ivation
    -1.48
    yline
    -1.44
    haven
    -1.43
    象
    -1.42
    POSITIVE LOGITS
    fold
    2.72
     dozen
    2.27
     hundred
    2.16
     thousand
    2.07
    leg
    2.03
     sets
    2.02
     successive
    1.87
     consecutive
    1.85
    thirds
    1.80
    hundred
    1.74
    Activations Density 0.168%

    No Known Activations