© 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. 25110
    Prev
    Next
    INDEX
    Explanations

    the word "while."

    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
    216
    +0.16
    0.9%
    273
    +0.15
    0.9%
    95
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    104
    +0.16
    0.06
    106
    +0.15
    0.06
    496
    +0.12
    0.06
    Negative Logits
     bench
    -1.86
    poons
    -1.82
    tti
    -1.68
    nement
    -1.67
     sentiment
    -1.67
    poon
    -1.63
     ridic
    -1.59
    ly
    -1.57
     criticism
    -1.54
    theless
    -1.51
    POSITIVE LOGITS
    ¦
    3.57
    Īĺ
    3.38
    Ń
    3.22
    ³
    3.10
    Ħ
    3.08
    ·
    3.07
    ¥
    2.98
    ¾
    2.94
    «
    2.87
    Ĥ¬
    2.84
    Activations Density 0.048%

    No Known Activations