© 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 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. 12956
    Prev
    Next
    INDEX
    Explanations

    programming or coding-related syntax and structures

    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
    66
    +0.12
    0.7%
    161
    +0.12
    0.7%
    378
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    20
    +0.12
    0.08
    66
    +0.12
    0.05
    264
    +0.11
    0.05
    Negative Logits
     Lithuan
    -1.96
     Natal
    -1.72
    ski
    -1.58
     Ce
    -1.54
     Serbia
    -1.52
     Bel
    -1.46
     Belgium
    -1.43
    ync
    -1.41
     aval
    -1.38
    zzi
    -1.36
    POSITIVE LOGITS
     H
    4.28
    H
    3.89
    HH
    3.40
    Hs
    3.31
     h
    3.29
    HB
    3.16
    HC
    3.15
     HH
    3.11
     HC
    3.10
    h
    3.09
    Activations Density 1.964%

    No Known Activations