© 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
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 4-RES-SM
    11. 11689
    Prev
    Next
    INDEX
    Explanations

    terms related to financial transactions and household utilities

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    348
    +0.22
    1.3%
    306
    +0.13
    0.8%
    466
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    154
    +0.22
    0.09
    348
    +0.13
    0.06
    137
    +0.12
    0.10
    Negative Logits
    ĥ½
    -1.60
    hift
    -1.53
    wards
    -1.46
    neath
    -1.43
    yon
    -1.42
    ubottu
    -1.41
    ounger
    -1.38
    ^)
    -1.37
     otherwise
    -1.37
    acer
    -1.34
    POSITIVE LOGITS
     etc
    2.78
     installations
    1.69
     or
    1.61
     towers
    1.61
     industries
    1.60
     machines
    1.59
     contests
    1.56
     disorders
    1.55
     commissions
    1.50
     monitors
    1.49
    Activations Density 1.166%

    No Known Activations