© 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. 9588
    Prev
    Next
    INDEX
    Explanations

    structured data and programming syntax

    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
    111
    +0.58
    4.8%
    156
    +0.51
    4.2%
    47
    +0.13
    1.1%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.58
    0.36
    156
    +0.51
    0.30
    71
    +0.13
    0.29
    Negative Logits
    wear
    -1.32
     figures
    -1.24
    âĢ¦”
    -1.23
    uate
    -1.22
    borough
    -1.21
     ‘
    -1.20
    amid
    -1.19
    nement
    -1.19
    uated
    -1.18
    icals
    -1.18
    POSITIVE LOGITS
    1.29
    #
    1.26
    1.26
    rosis
    1.26
     bool
    1.20
    1.19
     unicode
    1.19
     backend
    1.18
     stack
    1.16
     hash
    1.15
    Activations Density 0.399%

    No Known Activations