© 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. 26116
    Prev
    Next
    INDEX
    Explanations

    references to the word "wash."

    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
    +0.15
    0.9%
    125
    +0.15
    0.8%
    376
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    125
    +0.15
    0.04
    313
    +0.15
    0.04
    424
    +0.14
    0.03
    Negative Logits
    ances
    -1.60
    MENT
    -1.59
    edin
    -1.58
    izumab
    -1.54
    ments
    -1.53
    oise
    -1.52
    esis
    -1.51
    ancer
    -1.51
    estion
    -1.47
     entry
    -1.43
    POSITIVE LOGITS
    ither
    1.83
    avy
    1.73
    iring
    1.67
    abi
    1.63
    hered
    1.60
    ylan
    1.57
    avelength
    1.54
    iked
    1.51
    izards
    1.50
    addle
    1.44
    Activations Density 0.015%

    No Known Activations