© 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. 30246
    Prev
    Next
    INDEX
    Explanations

    instances of the word "then."

    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
    189
    +0.12
    0.7%
    428
    +0.11
    0.6%
    368
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    178
    +0.12
    0.06
    255
    +0.11
    0.06
    107
    +0.10
    0.06
    Negative Logits
    uary
    -1.75
    zes
    -1.50
    west
    -1.44
    arts
    -1.44
    ries
    -1.42
    °
    -1.41
    ships
    -1.39
    EP
    -1.39
    rics
    -1.38
     miles
    -1.37
    POSITIVE LOGITS
     ligation
    1.53
    ère
    1.48
     itself
    1.47
    otropic
    1.44
    ogenicity
    1.42
    .^\[[@
    1.41
    ité
    1.40
     given
    1.39
    ucleotide
    1.39
     extinction
    1.36
    Activations Density 0.055%

    No Known Activations