© 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 EvalsExportsAPI 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. 28783
    Prev
    Next
    INDEX
    Explanations

    instances of the word "supp."

    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
    362
    +0.16
    0.9%
    277
    +0.14
    0.8%
    424
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    277
    +0.16
    0.02
    424
    +0.14
    0.02
    480
    +0.12
    0.01
    Negative Logits
     toward
    -1.93
    Ļª
    -1.84
     how
    -1.65
     pursuant
    -1.58
     natural
    -1.57
     therefore
    -1.50
     nevertheless
    -1.47
    upon
    -1.46
    .’”
    -1.45
     consequences
    -1.45
    POSITIVE LOGITS
    houses
    1.60
    chap
    1.60
    journals
    1.54
    elist
    1.53
    skins
    1.50
    uild
    1.49
    gels
    1.49
    orters
    1.48
     patents
    1.48
    etable
    1.47
    Activations Density 0.006%

    No Known Activations