© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 25755
    Prev
    Next
    INDEX
    Explanations

    references to toxic substances and heavy metals

    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
    369
    +0.17
    1.0%
    175
    +0.12
    0.7%
    249
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    71
    +0.17
    0.01
    276
    +0.12
    0.01
    175
    +0.11
    0.01
    Negative Logits
     quer
    -1.85
    nered
    -1.56
    hes
    -1.52
     entire
    -1.46
    matically
    -1.39
     full
    -1.38
    ESULT
    -1.38
     Anyway
    -1.36
     broken
    -1.34
    mes
    -1.33
    POSITIVE LOGITS
     ions
    1.68
    assic
    1.51
    arman
    1.48
    thouse
    1.47
     industry
    1.46
    forge
    1.46
    istar
    1.43
     embargo
    1.43
    ocide
    1.42
    ilate
    1.41
    Activations Density 0.008%

    No Known Activations