© 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. 25199
    Prev
    Next
    INDEX
    Explanations

    nouns and pronouns

    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
    123
    +0.13
    0.8%
    219
    +0.11
    0.7%
    367
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    59
    +0.13
    0.01
    320
    +0.11
    0.01
    367
    +0.11
    0.01
    Negative Logits
     )
    -1.50
     possible
    -1.50
     Television
    -1.49
    ureus
    -1.48
     Teachers
    -1.47
     Sports
    -1.47
     Baltimore
    -1.45
    ?**
    -1.42
     Students
    -1.41
    '?"
    -1.41
    POSITIVE LOGITS
    making
    2.23
    cias
    2.09
    IFIC
    1.86
    isor
    1.72
    fest
    1.72
    finder
    1.67
    fiddle
    1.66
    storm
    1.57
    FW
    1.56
    makers
    1.53
    Activations Density 0.029%

    No Known Activations