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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 0-MLP-SM
    11. 30854
    Prev
    Next
    INDEX
    Explanations

    references to players in various contexts

    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
    156
    +0.17
    1.0%
    487
    +0.15
    0.9%
    376
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    487
    +0.17
    0.04
    329
    +0.15
    0.03
    112
    +0.15
    0.03
    Negative Logits
    woke
    -1.80
    ľĵ
    -1.74
    assing
    -1.65
    ONS
    -1.62
    ĥ
    -1.62
    °
    -1.61
    ras
    -1.61
    ennes
    -1.60
    ĥ½
    -1.60
    alus
    -1.58
    POSITIVE LOGITS
    hell
    1.90
    cript
    1.90
    creen
    1.66
    ystems
    1.64
    iative
    1.61
    uits
    1.59
    clerosis
    1.56
    hole
    1.55
    pieces
    1.53
    cribed
    1.52
    Activations Density 0.088%

    No Known Activations