© 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. 30967
    Prev
    Next
    INDEX
    Explanations

    references to the concept of "it" in various contexts

    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/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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    400
    +0.15
    0.9%
    55
    +0.15
    0.8%
    332
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    352
    +0.15
    0.20
    154
    +0.15
    0.17
    113
    +0.13
    0.15
    Negative Logits
     done
    -1.66
     anonymous
    -1.46
     Nations
    -1.41
     earlier
    -1.40
     anonymity
    -1.37
    NN
    -1.36
     scare
    -1.34
     NIH
    -1.32
     anonym
    -1.32
     powered
    -1.30
    POSITIVE LOGITS
    ()[
    1.90
    ENTER
    1.67
    ľ
    1.58
    iler
    1.56
    ented
    1.55
    etter
    1.52
    erp
    1.50
    rias
    1.46
    cal
    1.46
    clud
    1.45
    Activations Density 1.183%

    No Known Activations