© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 0-MLP-SM
    11. 30879
    Prev
    Next
    INDEX
    Explanations

    mentions of anniversaries and significant celebrations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.0.hook_mlp_out
    Hook Layer
    0
    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
    30
    +0.11
    0.6%
    122
    +0.11
    0.6%
    376
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    122
    +0.11
    0.01
    212
    +0.11
    0.01
    30
    +0.11
    0.01
    Negative Logits
    UTH
    -1.63
    rug
    -1.59
     any
    -1.49
    Īĺ
    -1.48
     lesbian
    -1.44
     both
    -1.44
    )</
    -1.43
     placebo
    -1.41
    unde
    -1.41
     happier
    -1.40
    POSITIVE LOGITS
    opan
    1.85
    umes
    1.60
     celebrations
    1.57
    ume
    1.57
     dimensions
    1.54
     positions
    1.53
    ios
    1.52
    raits
    1.52
    pieces
    1.52
    picker
    1.51
    Activations Density 0.391%

    No Known Activations