© 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. 29737
    Prev
    Next
    INDEX
    Explanations

    terminology related to degenerative diseases

    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
    156
    +0.23
    1.3%
    376
    +0.17
    1.0%
    446
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    325
    +0.23
    0.01
    266
    +0.17
    0.01
    259
    +0.11
    0.01
    Negative Logits
    ows
    -1.59
    ku
    -1.52
    kie
    -1.49
    ikh
    -1.48
     Available
    -1.44
    glich
    -1.35
    eps
    -1.33
    ivalent
    -1.33
     publish
    -1.31
     "")
    -1.30
    POSITIVE LOGITS
    isms
    1.65
    freak
    1.49
    guy
    1.45
     asshole
    1.42
    istically
    1.42
     rumors
    1.40
    ism
    1.40
     wrongdoing
    1.38
    "}](#
    1.38
     mistakes
    1.34
    Activations Density 0.008%

    No Known Activations