© 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. 26898
    Prev
    Next
    INDEX
    Explanations

    instances of the name "Joe" or variations of "Jesuit"

    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
    410
    +0.15
    0.9%
    369
    +0.13
    0.7%
    111
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.15
    0.01
    410
    +0.13
    0.01
    375
    +0.13
    0.01
    Negative Logits
    ){#
    -1.67
     Russia
    -1.63
     Serbia
    -1.48
     distance
    -1.48
    rapy
    -1.43
     important
    -1.42
    noreply
    -1.42
     Moscow
    -1.42
    tering
    -1.40
     Vladimir
    -1.38
    POSITIVE LOGITS
    veau
    2.00
    été
    1.85
    illard
    1.83
    ilee
    1.71
    ÅĽci
    1.69
    arty
    1.69
    chitz
    1.66
    ptions
    1.63
    cript
    1.62
    ÅĦ
    1.59
    Activations Density 0.014%

    No Known Activations