© 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. Residuals
    9. 1-RES-SM
    11. 24353
    Prev
    Next
    INDEX
    Explanations

    references to pop culture and its various aspects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/1-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.1.hook_resid_post
    Hook Layer
    1
    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
    164
    +0.11
    0.6%
    369
    +0.10
    0.6%
    410
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    509
    +0.11
    0.02
    452
    +0.10
    0.02
    437
    +0.10
    0.02
    Negative Logits
     slightest
    -1.79
     plain
    -1.49
     NHS
    -1.45
     Command
    -1.41
     swift
    -1.40
     same
    -1.38
     certain
    -1.34
    documentclass
    -1.33
     leaking
    -1.28
     expected
    -1.26
    POSITIVE LOGITS
    lite
    2.45
    ulates
    2.12
    ups
    2.03
    corn
    2.03
    culo
    1.99
    ulous
    1.85
    starter
    1.84
    ipers
    1.84
    holder
    1.80
    holders
    1.72
    Activations Density 0.116%

    No Known Activations