© 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. 22465
    Prev
    Next
    INDEX
    Explanations

    references to religious figures, authority, and church hierarchies

    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/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
    Neuron Alignment
    Index
    Value
    % of L₁
    156
    +0.26
    1.5%
    426
    +0.16
    0.9%
    260
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    426
    +0.26
    0.07
    122
    +0.16
    0.05
    339
    +0.11
    0.09
    Negative Logits
    ŀ
    -1.77
    °
    -1.63
    олÑĮ
    -1.51
     0000000000000000000000000000000000
    -1.46
    [\#
    -1.45
    geant
    -1.44
     met
    -1.40
    idone
    -1.35
    ener
    -1.34
    ppat
    -1.34
    POSITIVE LOGITS
    iness
    1.68
    izations
    1.66
    ages
    1.61
     yourselves
    1.54
    ories
    1.53
     yourself
    1.52
     researchers
    1.51
    acity
    1.49
    oretical
    1.48
    ities
    1.45
    Activations Density 1.989%

    No Known Activations