© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 0-MLP-SM
    6. 27082
    Prev
    Next
    INDEX
    Explanations

    the word "both" indicating the presence of duality or comparison

    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
    139
    +0.11
    0.6%
    334
    +0.10
    0.6%
    220
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    353
    +0.11
    0.07
    357
    +0.10
    0.06
    207
    +0.10
    0.06
    Negative Logits
    undry
    -1.43
    documentclass
    -1.37
    less
    -1.36
    boolean
    -1.34
    int
    -1.32
    ighter
    -1.31
    esto
    -1.31
     reminder
    -1.29
     alike
    -1.28
    ella
    -1.27
    POSITIVE LOGITS
     sexes
    2.40
     sides
    1.73
    iances
    1.59
    etine
    1.59
    urope
    1.46
    rops
    1.41
    ystems
    1.38
     births
    1.36
    chitz
    1.34
    obar
    1.33
    Activations Density 0.095%

    No Known Activations