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. 27520
    Prev
    Next
    INDEX
    Explanations

    special character usage in code or expressions

    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
    23
    +0.37
    2.2%
    250
    +0.16
    1.0%
    321
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    250
    +0.37
    0.03
    321
    +0.16
    0.02
    247
    +0.15
    0.02
    Negative Logits
    aho
    -1.73
    erior
    -1.72
    ruptcy
    -1.70
    zol
    -1.64
    ÅĽ
    -1.60
    bie
    -1.60
    writer
    -1.57
    umab
    -1.57
    etry
    -1.57
    heed
    -1.56
    POSITIVE LOGITS
    oved
    1.50
    uously
    1.49
    ipt
    1.47
    EVER
    1.44
     settled
    1.44
     footsteps
    1.43
     ways
    1.42
     real
    1.42
    nt
    1.42
    athered
    1.41
    Activations Density 0.043%

    No Known Activations