© 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. 31034
    Prev
    Next
    INDEX
    Explanations

    instances of the word "work"

    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
    286
    +0.14
    0.8%
    70
    +0.13
    0.8%
    80
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    80
    +0.14
    0.02
    159
    +0.13
    0.02
    286
    +0.12
    0.01
    Negative Logits
    Ĩ
    -1.80
    ľ
    -1.71
    uran
    -1.65
    apers
    -1.65
    ureus
    -1.55
    ppers
    -1.55
    ursed
    -1.54
     say
    -1.54
    ĥ½
    -1.49
    rians
    -1.46
    POSITIVE LOGITS
    enstein
    2.15
    yard
    1.95
    die
    1.92
    bare
    1.74
    este
    1.71
    house
    1.69
    holm
    1.68
    dan
    1.67
    haus
    1.64
    bilt
    1.64
    Activations Density 0.421%

    No Known Activations