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. 27566
    Prev
    Next
    INDEX
    Explanations

    phrases emphasizing financial details

    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
    48
    +0.12
    0.7%
    460
    +0.11
    0.6%
    355
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    98
    +0.12
    0.17
    23
    +0.11
    0.34
    321
    +0.11
    0.21
    Negative Logits
    ippy
    -1.62
    bounds
    -1.51
    atives
    -1.45
    obb
    -1.42
     Amsterdam
    -1.39
    vre
    -1.38
    que
    -1.38
    ^-
    -1.37
    ibilities
    -1.35
     Barcelona
    -1.34
    POSITIVE LOGITS
    !”
    1.76
    ¦
    1.76
    »
    1.75
    ĩ
    1.68
    Ľ
    1.62
    ¼
    1.61
    ±
    1.60
    unreadable
    1.59
    sis
    1.58
    ¾
    1.49
    Activations Density 4.829%

    No Known Activations