© 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. 25609
    Prev
    Next
    INDEX
    Explanations

    contractions, specifically "won't" and "don't"

    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
    250
    +0.14
    0.8%
    478
    +0.12
    0.7%
    372
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    250
    +0.14
    0.08
    372
    +0.12
    0.08
    147
    +0.12
    0.07
    Negative Logits
    ously
    -1.64
    anic
    -1.62
    uld
    -1.49
    istine
    -1.47
    ism
    -1.45
     manager
    -1.37
    ampoline
    -1.37
    struct
    -1.35
     weren
    -1.35
    anced
    -1.33
    POSITIVE LOGITS
    orge
    1.95
    RON
    1.70
    ablo
    1.58
    Appeal
    1.57
    CoV
    1.57
    jä
    1.50
     anymore
    1.50
    iq
    1.47
    že
    1.46
    reon
    1.46
    Activations Density 0.254%

    No Known Activations