© 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. 12243
    Prev
    Next
    INDEX
    Explanations

    references to background or contextual information

    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
    216
    +0.15
    0.9%
    377
    +0.13
    0.7%
    16
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    216
    +0.15
    0.01
    16
    +0.13
    0.01
    48
    +0.12
    0.01
    Negative Logits
    abase
    -1.65
     favor
    -1.60
    rying
    -1.60
    ican
    -1.56
    thon
    -1.50
    hetti
    -1.50
     seriously
    -1.48
    ials
    -1.48
    inely
    -1.42
     challenged
    -1.40
    POSITIVE LOGITS
     Facts
    2.10
     Art
    1.83
     {#
    1.74
     subtraction
    1.69
     Description
    1.67
     Reduction
    1.65
     Affairs
    1.61
    method
    1.60
     art
    1.56
     Characteristics
    1.54
    Activations Density 0.015%

    No Known Activations