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. 27442
    Prev
    Next
    INDEX
    Explanations

    occurrences of the number 24

    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
    +0.17
    1.0%
    458
    +0.15
    0.9%
    477
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    458
    +0.17
    0.04
    445
    +0.15
    0.04
    498
    +0.13
    0.04
    Negative Logits
    rition
    -1.65
     Wars
    -1.63
     possession
    -1.62
     World
    -1.60
     Circuit
    -1.54
     sale
    -1.52
     responsibility
    -1.50
     world
    -1.46
     parap
    -1.44
     title
    -1.41
    POSITIVE LOGITS
    ³
    3.41
    ĥ½
    3.24
    ª
    3.23
    µ
    3.11
    ®
    3.08
    ·¸
    3.06
    Ń
    2.88
    ħ
    2.85
    ¬
    2.85
    ¾
    2.85
    Activations Density 0.032%

    No Known Activations