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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 0-MLP-SM
    6. 28027
    Prev
    Next
    INDEX
    Explanations

    references to potatoes and related products

    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
    +0.13
    0.7%
    330
    +0.10
    0.6%
    43
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    168
    +0.13
    0.01
    330
    +0.10
    0.01
    416
    +0.09
    0.01
    Negative Logits
     mean
    -1.68
     frequent
    -1.62
     your
    -1.60
     yours
    -1.55
    ird
    -1.55
    rian
    -1.54
    '));
    -1.52
     happier
    -1.48
     their
    -1.48
     median
    -1.45
    POSITIVE LOGITS
    vill
    1.95
    ÅĽci
    1.94
    vä
    1.88
    ville
    1.80
     defense
    1.78
    defense
    1.67
    castle
    1.63
    Vill
    1.61
    asek
    1.59
    works
    1.58
    Activations Density 0.013%

    No Known Activations