Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 0-MLP-SM
    6. 27430
    Prev
    Next
    INDEX
    Explanations

    the word "Tang" and its variations or associations with "Ming."

    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
    369
    +0.18
    1.0%
    203
    +0.14
    0.8%
    81
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    332
    +0.18
    0.01
    59
    +0.14
    0.01
    71
    +0.12
    0.01
    Negative Logits
    Ļª
    -1.61
     tracing
    -1.60
    Ĥ¬
    -1.52
     marker
    -1.50
    comple
    -1.48
     masking
    -1.40
     correcting
    -1.37
     covering
    -1.36
    ção
    -1.36
     connecting
    -1.36
    POSITIVE LOGITS
    urd
    1.99
    doms
    1.94
    ultan
    1.80
    iei
    1.75
    ued
    1.72
    iani
    1.72
    ief
    1.71
    ue
    1.68
    eni
    1.67
    iev
    1.65
    Activations Density 0.023%

    No Known Activations