© 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. 25173
    Prev
    Next
    INDEX
    Explanations

    instances of strong emotional language related to guilt and punishment

    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
    462
    +0.18
    1.0%
    156
    +0.17
    1.0%
    10
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    203
    +0.18
    0.36
    23
    +0.17
    0.31
    494
    +0.14
    0.25
    Negative Logits
    ±
    -1.96
    ↵
    -1.84
    -1.84
    -1.84
    ↵↵
    -1.84
    ↵
    -1.84
    -1.84
    ↵
    -1.84
    ↵
    -1.84
          ↵
    -1.84
    POSITIVE LOGITS
    hin
    1.60
    >/
    1.57
    ols
    1.57
    ney
    1.50
    ters
    1.47
    ette
    1.44
    nis
    1.43
     \!
    1.42
    ico
    1.40
     Caption
    1.39
    Activations Density 3.097%

    No Known Activations