© 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. 2-RES-SM
    11. 13023
    Prev
    Next
    INDEX
    Explanations

    specific identifiers and metadata related to scientific articles and studies

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    271
    +0.14
    0.8%
    156
    +0.12
    0.7%
    315
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    271
    +0.14
    -0.01
    445
    +0.12
    0.01
    315
    +0.12
    0.01
    Negative Logits
    ."[
    -1.47
     redundancy
    -1.41
     preserved
    -1.35
     remains
    -1.35
     Trustees
    -1.32
     retained
    -1.31
     Norfolk
    -1.30
     Danish
    -1.29
     Delaware
    -1.27
     PCB
    -1.26
    POSITIVE LOGITS
    IJ
    2.59
    ĵ
    2.48
    Ĵ
    2.41
    ı
    2.31
    Ļª
    2.18
    »¿
    2.15
    ģ
    2.15
    ł
    2.07
    Ĩ
    2.03
    Ŀ
    2.00
    Activations Density 0.427%

    No Known Activations