© 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. 2-MLP-SM
    11. 22670
    Prev
    Next
    INDEX
    Explanations

    mathematical and programming expressions involving variables and calculations

    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/2-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.2.hook_mlp_out
    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
    156
    +0.24
    1.4%
    507
    +0.12
    0.7%
    23
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    187
    +0.24
    0.01
    400
    +0.12
    0.04
    23
    +0.12
    0.03
    Negative Logits
    heets
    -1.53
    guard
    -1.41
    ags
    -1.35
    sek
    -1.34
    atts
    -1.34
     gears
    -1.34
     additions
    -1.33
     imagin
    -1.29
    peg
    -1.28
     ges
    -1.25
    POSITIVE LOGITS
    ality
    1.90
    purpose
    1.64
     convenience
    1.57
     Abuse
    1.53
    alities
    1.52
    collapse
    1.49
     benefit
    1.45
    thood
    1.44
     purpose
    1.43
     massacre
    1.41
    Activations Density 1.557%

    No Known Activations