© 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. 25592
    Prev
    Next
    INDEX
    Explanations

    phrases expressing the concept of fitness or suitability for a particular purpose

    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
    71
    +0.15
    0.9%
    251
    +0.13
    0.8%
    143
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    143
    +0.15
    0.02
    251
    +0.13
    0.02
    65
    +0.13
    0.02
    Negative Logits
    ---|---|---
    -1.67
    rake
    -1.56
    tox
    -1.53
     Peace
    -1.42
     peace
    -1.38
     deployment
    -1.36
     safely
    -1.36
    utz
    -1.36
     Massachusetts
    -1.34
     |
    -1.33
    POSITIVE LOGITS
    IFIC
    1.84
    heets
    1.81
    ific
    1.73
    krit
    1.71
    oslov
    1.62
    cards
    1.61
    fonts
    1.60
    manship
    1.59
    naire
    1.57
    aho
    1.55
    Activations Density 0.131%

    No Known Activations