© 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 EvalsExportsAPI 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. 27874
    Prev
    Next
    INDEX
    Explanations

    instances of the phrase "to" followed by an action or verb indicating intention or 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
    188
    +0.14
    0.8%
    45
    +0.11
    0.6%
    439
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    69
    +0.14
    0.38
    434
    +0.11
    0.30
    151
    +0.11
    0.31
    Negative Logits
    opically
    -1.78
    olars
    -1.49
    rieved
    -1.47
     versus
    -1.47
    ear
    -1.47
     classified
    -1.46
    acial
    -1.45
     lasts
    -1.44
     million
    -1.43
    az
    -1.42
    POSITIVE LOGITS
     Parish
    1.86
    chard
    1.82
    iginally
    1.66
    LINE
    1.52
    forth
    1.51
     Indies
    1.49
     whatever
    1.47
     hereinafter
    1.45
    sey
    1.36
     offence
    1.35
    Activations Density 1.660%

    No Known Activations