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. 27497
    Prev
    Next
    INDEX
    Explanations

    instances of the string "STRING."

    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
    +0.14
    0.8%
    376
    +0.14
    0.8%
    334
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    334
    +0.14
    0.01
    232
    +0.14
    0.01
    263
    +0.13
    0.01
    Negative Logits
    cludes
    -1.62
     true
    -1.59
     href
    -1.58
    cluding
    -1.58
     immediate
    -1.58
    acity
    -1.56
    sis
    -1.54
     Commons
    -1.46
    stitutional
    -1.45
    gress
    -1.42
    POSITIVE LOGITS
    ings
    1.97
    ONG
    1.93
    ument
    1.68
    INGS
    1.67
    \.
    1.61
    ouin
    1.61
    asso
    1.57
    ingly
    1.56
    UMENT
    1.56
    \*,
    1.55
    Activations Density 0.006%

    No Known Activations