© 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. 25194
    Prev
    Next
    INDEX
    Explanations

    references to geographical regions, specifically Europe and America

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    481
    +0.15
    0.8%
    182
    +0.15
    0.8%
    478
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.15
    0.04
    71
    +0.15
    0.02
    321
    +0.14
    0.02
    Negative Logits
     (\<
    -1.54
     clue
    -1.54
     resemblance
    -1.38
     interface
    -1.38
     clips
    -1.36
     compilation
    -1.36
     Commod
    -1.33
     Regulations
    -1.32
     luck
    -1.30
     hints
    -1.29
    POSITIVE LOGITS
    §
    3.27
    ©
    2.96
    Ń
    2.55
    ¯
    2.54
    ±
    2.51
    µ
    2.44
    Ĭ
    2.42
    ŀ
    2.42
    ½
    2.41
    ¨
    2.41
    Activations Density 0.087%

    No Known Activations